Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

A traditional agroforestry landscape of Ferguson Island, Papua New Guinea Flavelle, Alix J. 1991

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1991_A6_7 F62.pdf [ 10.16MB ]
JSON: 831-1.0098572.json
JSON-LD: 831-1.0098572-ld.json
RDF/XML (Pretty): 831-1.0098572-rdf.xml
RDF/JSON: 831-1.0098572-rdf.json
Turtle: 831-1.0098572-turtle.txt
N-Triples: 831-1.0098572-rdf-ntriples.txt
Original Record: 831-1.0098572-source.json
Full Text

Full Text

A TRADITIONAL AGROFORESTRY LANDSCAPE ON FERGUSSON ISLAND, PAPUA NEW GUINEA by Alix J . Flavelle B.A. Biology/Philosophy, University of Victoria, 1 9 8 3 A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in The Faculty of Graduate Studies (Department of Forest Resources Management) We accept this thesis as conforming to the required standard The University of British Columbia September 1 9 9 1 Q Alix J . Flavelle 1 9 9 1 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of \-Qyz<>\ QeoQQ/tizf McUn&^wei^J' The University of British Columbia Vancouver, Canada Date DE-6 (2/88) ABSTRACT A study of a traditional land use system was conducted at Nade, Fergusson Island, in Milne Bay Province, Papua New Guinea. An ethnobotanical inventory of useful and culturally significant plants, and a series of transects and vegetation profiles were used to identify, and map the distribution of, 15 different plant communities in the Nade landscape. Interviews were conducted with local gardeners about land use decision-making, land tenure, and ecological knowledge. The land use strategy practiced at Nade can be characterized as a polyphase agroforestry system. A spectrum of management techniques are used in the different phases, including the selecting, ignoring, transplanting and/or planting of wild, semi-domesticated, and domesticated tree species. A variety of subsistence products are available throughout the year, from the range of vegetation types. The distribution of successional phases in the landscape was found to depend on topography and soil conditions which vary within the subsistence territory of Nade. Overlying the environmentally determined pattern of the shifting mosaic are the social factors; land use decision-making based on the traditional system of susu land and plant tenure, labour-saving strategies, and agricultural tradition. The study provides baseline data for monitoring changes in the culturally modified landscapes of Fergusson Island. This in turn can be used to facilitate a land-use planning process with local people. i i TABLE OF CONTENTS Abstract ii Table of Contents iii List of Figures vi List of Tables vii List of Maps viii List of Appendices ix Acknowledgements '. x Chapter 1. INTRODUCTION 1 Chapter 2. APPROACHES TO THE STUDY OF TRADITIONAL RAINFOREST MANAGEMENT AND AGROFORESTRY SYSTEMS: A SURVEY 5 2.1 Approaches to the study of traditional land use systems in tropical rainforest 5 2.2 Characterization of traditional agroforestry systems 6 2.2.1 Polyculture and monoculture 6 2.2.2 Polyculture and polyvariety 8 2.2.3 Polyphase agroforestry systems 8 2.2.4 Characterizing horticultural techniques 9 2.3 Traditional exploitation of Melanesian forest ecosystems 11 2.3.1 Agroforestry systems of Melanesia 13 2.3.2 Human impacts on the Melanesian environment 13 Chapter 3. METHODOLOGY 15 3.1 .Foundational Questions 15 3.2 Methodology 16 3.2.1 Ethnobotanical inventory 18 3.2.2 Maps and transects 19 3.2.3 Vegetation profiles 19 3.2.4 Garden diagrams 20 3.2.5 Interview techniques 20 3.2.6 Land tenure 21 3.2.7 Schedule of research activities 21 Chapter 4. A BIOPHYSICAL DESCRIPTION OF FERGUSSON ISLAND 23 4.1 Location 23 4.2 Geology and geomorphology 23 4.3 Soils 24 4.4 Climate 24 4.5 Biogeography and ecology 24 Chapter 5. THE HISTORICAL AND CULTURAL CONTEXT 28 5.1 The Massim 28 5.2 European influence 30 5.3 The establishment of Nade 31 5.4 Nade social organization 32 5.5 Exchange - the Massim perspective on land and food 33 Chapter 6. PLANT INVENTORY 36 6.1 The importance of plants in the Nade subsistence economy 36 6.2 Plants cultivated in the swiddens 38 6.3 Multi-use plants 38 6.4 Food plants 39 6.5 Construction plants 40 6.6 Fibre plants 40 6.7 Medicinal plants and magic plants 41 6.8 Ornamental plants 43 6.9 Plants of ecological significance 43 Chapter 7. CLASSIFICATION, DISTRIBUTION, AND MANAGEMENT OF SUCCESSIONAL PHASES IN THE NADE LANDSCAPE 46 7.1 Classification of successional phases in the Nade cultural landscape ... 46 7.2 The influence of environmental parameters in the Nade landscape 65 7.3 The influence of management techniques on the Nade landscape 67 7.4 The agroforestry landscape at Nade.; 72 Chapter 8. THE SOCIAL ORGANIZATION OF LAND: CUSTOMARY LAND TENURE, AND LAND USE DECISION-MAKING.... 77 8.1 The customary land tenure system 77 8.1.1 The land tenure system 77 8.1.2 The plant tenure system 80 8.1.3 Sources of tenure disputes 81 8.1.4 The resolution of tenure disputes 83 8.2 Vectors of change in the customary tenure system 83 8.3 Sacred limitations on cultivation 86 8.4 Land use decision-making 88 8.5 The cultural influence on the Nade landscape 90 Chapter 9. FOREST RESOURCE MANAGEMENT AT NADE: OPPORTUNITIES AND CONSTRAINTS OF THE TRADITIONAL AGROFORESTRY SYSTEM 93 9.1 The development context in Papua New Guinea 93 9.1.1 Forest development in Papua New Guinea : 93 9.1.2 Customary laws and land tenure 96 9.1.3 Conservation area planning 98 9.2 The potential for ecodevelopment on Fergusson Island 101 9.2.1 Conservation and ecotourism planning 102 9.2.2 Adapting traditional agroforestry 105 Chapter 10. CONCLUSION 111 Footnotes 113 Bibliography 117 Appendices 131 v FIGURES Figure 1. Seasonal plant availability Figure 2. Schematic diagram of transect of study area Figure 3. Schematic diagram of a one year old .... Figure 4. Schematic diagram of a tala swidden garden cluster Figure 5. Schematic profile of hamlet agroforest Figure 6. Schematic profile of mixed sago grove Figure 7. Schematic profile of fallow phase....... Figure 8. Schematic profile of modified lowland forest Figure 9. Schematic profile of modified small-crowned hill forest. LIST OF TABLES Table B l . Ethnobotanical inventory 132 Table B2. Vernacular list of plants of undeteirnined use 161 Table CI. Summary of swidden garden data 163 Table C2 Summary of Vegetation profiles 166 v i i MAPS Map 1. D'Entrecasteaux Islands, Milne Bay Province, South Eastern Papua New Guinea. 1:600,000 4 Map 2. Topographic features and soils. 1:35,000 27 Map 3. Present hamlet locations. 1:35,000 35 Map 4. Distribution of 1989 and 1990 swiddens. 1:35,000 49 Map 5. Distribution of successional phases of the Nade agroforestry 50 landscape. 1:35,000 Map 6. Cultural features of the Nade landscape. 1:35,000. 87 Map 7. Map of survey plot locations. 1:35,000 181 8 The Nade cultural setting: settlements and historical sites. 1:10,000 Map 9. Distribution of swiddens and successional phases of the Nade landscape. 1:10,000 v i i i APPENDICES Appendix A. Note on folk classification 131 Appendix B. Ethnobotanical inventory 132 Appendix C. Vegetation profiles and swidden data 163 Appendix D. Study location 181 Appendix E. Questionnaire 182 Appendix F. Map 8 The Nade cultural landscape: settlements, graveyards, and sacred sites. Map 9 Distribution of swidden and successional phases in the Nade agroforestry landscape. Standards used in this text. t When refered to in this thesis, names in the Nade language will be italicized, latin botanical names will be underlined, and common names will appear in regular text. i x ACKNOWLEDGEMENTS The efforts of many people have made this thesis possible. I would first like to thank the people of Nade village on Fergusson Island for their hospitality, and for sharing with me some of their stories, their laughter, their knowledge, and a small segment of their lives. Many people in Papua New Guinea made the field research possible: Lance Hill and Helen Hopkins of the University of Papua New Guinea, Biology Department; Jacob Simet and Oswald Opaivu of the PNG Institute of Papua New Guinea Studies; Ian Fraser, UPNG, Dept. of Law; Karl Karenga and Osia Gideon at the PNG Forest Research Institute; and Sanori Elliot at the Milne Bay Province, Planning Department; William Wapili, Esa'Ala District forest officer. At UBC, I would like to thank Dr. Brent Ingram for his enthusiastic support and for introducing me to the international network of people working in tropical forestry and conservation. Dr. John Barker attempted the impossible task of preparing me for living in and learning about a Melanesian culture, and offered valuable suggestions in my attempt to integrate the disciplines of anthropology and forest management. Dr. Allan Chambers provided consistently wise advice in the research design and the writing of this thesis. I thank Dr. Jack Thirgood for some initial discussion of ideas on social aspects of forestry. I would like to thank the International Development Research Centre for financing the research. Finally, merci bien to my partner Marcel for his patience and his faithful support and encouragement throughout the whole process. x Chapter 1. INTRODUCTION Indigenous societies associated with tropical rainforests have developed various strategies for the exploitation of forest resources, some more successful and sustainable than others. Such strategies are adaptations to the opportunities and constraints of the broad range of tropical habitats. Both the environmental and the cultural diversity in the tropics would suggest that a variety of forest/agricultural land use systems exist. Yet characterization of these systems is often oversimplified and documentation is far from complete. Through research done in the fields of ethnobotany and human ecology, some commonalities in traditional agroecosystems in rainforest have become widely recognized. They are usually polycultural, often involving the management of wild plant and animal species and multiple crops cultivated together. When annual crops and trees are managed in combination the term agroforestry is applied. Such systems depend on renewable resources and generally require less external input, relying on ecological mechanisms for weed and insect control and maintenance of soil fertility (Alcorn 1990; Gleissman 1990; Janzen 1973). Finally, the systems are based on and regulated by local ecological knowledge meshed with social patterns and cultural beliefs. Traditional agroecosystems have generally evolved within the constraints of a certain resource base and population density. In recent decades however, commercial forestry and mining operations, the urgent need for established conservation areas, and growing human populations have placed increasing and conflicting demands upon the forest land and resources. These demands usually force intensification of traditional forest management systems. A common result is increased forest clearing by rainforest inhabitants, and shorter fallow periods, leading to soil degradation. At the same time, expansion of modern technologies and markets is changing human relationships with the forest. This often creates a trend toward using a lower diversity of subsistence crops as cash crops replace traditional forms of livelihood. And related to the real loss of plant diversity in the tropical rainforests 1 is that the knowledge about uses of plants is being lost as market economies and Western medical and educational systems are integrated by indigenous peoples (Balick 1990). Traditional agroforestry systems may offer viable alternatives to destructive forms of shifting cultivation, and clearcut logging of rainforests. Studies of traditional forest use can help determine to what extent these systems may be adapted to accomodate increasing material needs of rural communities in rainforest regions. And such studies can help to develop models for the conservation of tropical biodiversity, involving initiatives by, and co-management with, traditional communities. Integrating local skills, knowledge, and beliefs will help lay the basis for the cultural and political empowerment of rainforest communities which are being increasingly marginalized. Melanesia is an island region (including Papua New Guinea, Fiji, New Caledonia, Solomon Islands, Bismarck Archipelago, New Ireland, and New Hebrides) that is undergoing rapid political, social, economic, and environmental change. It is a region of cultural richness and variety, and diverse landscapes ranging from mountain glaciers to coral atolls, to mangrove swamps. The island ecosystems support a distinctive flora and fauna with high levels of endemism in limited habitats that are especially vulnerable to degradation. The development of island resources by forest clearing for agricultural and timber production has reduced many island habitats, particularly in coastal areas. Island Melanesia has one of the highest rates and numbers of species extinctions in the world (Dahl 1984). Fergusson Island, one of the largest islands of the D'Entrecasteaux Archipelago, which lies off the southeastern tip of the New Guinea mainland (Map 1), was chosen for this study for two primary reasons. First, the land-use systems practiced there are still traditional, though changes are foreseen in the near future due to the increasing population, commercial land-use pressures (logging and mining), and the transition from a local subsistence economy to a partially cash economy (Pernetta and Hill 1982). Second, the island is listed as a high priority for conservation programmes in the Review of the Protected Areas System of  Oceania (IUCN 1986). 2 The 15,000 inhabitants of Fergusson Island are still primarily dependent on their forest and marine-based subsistence economy. In the coming decades, however, industrial forestry and mining may diminish lands available for traditional exploitation. Observations made by Ingram, in 1987 and 1989, of intensified cropping and forest burning at lower elevations suggest that the limits of the human carrying capacity for the production of some species and crops may be close to being reached (G.B. Ingram, Dept. of Plant Science, UBC). The traditional forest-use strategy observed by Ingram (1989) includes swidden* gardens, intensively managed forest, and grasslands. Much of the island within 2 km of shore is under some form of traditional management; and this cultural landscape zone may be increasing with increasing population. There has been no research focused on traditional land-use strategies for the D'Entrecaseaux islands. This ethno-ecological study (Conklin 1957) describes and characterizes the traditional agroforestry landscape at the village of Nade, on Fergusson Island, on the basis of an inventory of the modified and natural successional plant community types, and locally used plant species. The data provides baseline information for monitoring change in the culturally-modified forest landscapes of Fergusson Island. This study, and the research priorities it identifies, could be used for long-term land use planning to facilitate the striking of a balance between resource development, habitat conservation, and traditional agroforestry on Fergusson Island. 3 9° 10' Lat. Cape Vogel 150°00'Long Amphlet Islands ^0 Scale 1:600,000 20km o Uama Island A 1947m \ Tewara Island Sanaroa Island Nade (« Budoya^ -Esa'ala//i ^JyDobu Island 1 N Map 1. D'Entrecasteaux Islands, Milne Bay Province, South Eastern Papua New Guinea. Chapter 2. APPROACHES TO THE STUDY OF TRADITIONAL RAINFOREST  MANAGEMENT AND AGROFORESTRY SYSTEMS: A SURVEY 2.1 Approaches to the study of traditional land use in tropical rainforest regions. Any traditional land use system depends on a complex and unique combination of environmental variables - the description of which can be an end in itself. But a land use system does not exist independently of the social organization of the people involved in it (Ellen 1982). The debate continues about cultural or environmental determinism in the evolution of subsistence strategies. But whether we seek theories of determinism and adaptation, or simply enough understanding to guide land use planning decisions, it is evident that the study of traditional land use requires an interdisciplinary approach. Methodology and analyses may come from such diverse disciplines as ecology, anthropology, botany, zoology, geography, and agronomy. Ecological studies by human geographers and social anthropologists became common in the 1970's, perhaps reflecting the interest in systems ecology and the strengthening environmental movement. Many of these studies focused on the flow of energy through small-scale societies (Rappaport 1967; Clarke 1971; Waddell 1972,) or on the ecological dynamics of cultivation in the rainforest (Posey 1985; Manner 1981). Ethnographers have described the cultural and social regulation of forest resource inputs and outputs of such systems (Rappaport 1967); and looked for patterns in the local ecological knowledge (Hayes 1983). When we say forest management our first thoughts lean towards the operations of a forester trained in some variation of the Western tradition - operations designed to extract wood that is mostly destined for developed countries. Yet forest management has been, and still is, in many areas, undertaken by local people without any academic training; people who perceive and utilize the environment differently; people who have managed their forests for millennia, and who have complex traditions that even now attract little scientific attention other than of an external and exploitive kind (Webb 1982:159). Research interest in agroforestry systems is new, but the farming systems are not, and may be more widespread than previously thought. Agroforestry, in the form of swidden cultivation, is the foundation of agriculture throughout the lowland tropics. While not all 5 swidden cultivation is agroforestry, most traditional systems purposely plant or protect woody perennials for subsequent harvest. As an ancient subsistence technique swiddening is often placed in a broad subsistence category and made to represent a complete system of livelihood. This hides important variations. More accurately, swiddening may be part of a complex strategy linking together a range of techniques (Ellen 1982: 170) in agriculture, silviculture, gathering, hunting, fishing, and animal husbandry. The combination of techniques plays a key role in the viability of the subsistence economy. Meeting a community's needs by swidden cultivation alone would often be too intense to be sustained in most environments (unless external inputs such as fertilizers and pesticides are used). 2.2 Characterization of traditional agroforestry systems 2.2.1 Polyculture or monoculture Polyculture refers to the systematic interplanting or tending of a variety of cultigens in a garden. Many researchers have suggested that polyculture is a common characteristic of traditional agroecosystems in tropical rainforest regions (Conklin 1957, Miracle 1967, Igbozurike 1971; Gleissman 1981), and that monoculture is typical of the modern, "artificial" agroecosystem (Janzen 1975; Meggers 1971). Traditional agroforestry is an agroecosystem involving a random mix of annual and perennial species, and in some cases, animals. Modern agroforestry designs involve the orderly interplanting of one or two multipurpose shade trees with one or two annual food crops. They do not represent the same degree of polyculture found in traditional agroforestry systems. The dichotomy of modern and traditional agroforestry does not actually exist. Some authors claim that polyculture is less typical of tropical rainforest horticulture than previously thought, and that monoculture with polyvariety is also common (Hames 1983; Clawson 1985). Furthermore, some traditional agroforesters are now combining cash crops in their subsistence cropping pattern. 6 The most studied examples of polycultural agroforestry are the homegardens found throughout the humid tropics, in Central America (Alcorn 1981; Mergen 1987), in Southeast Asia (Michon et al 1986; Soewartomo 1987), in South America (Ninez 1985; Padoch and de Jong 1991), and in Africa (Fernandez et al 1983) and in urban PNG (Vasey 1985). Homegardens are characterized as complexes of annual and perennial crops maintained near rural homes, which provide multiple products such as staple and snack foods, medicines, ornamentals, construction materials, and fibrecraft materials. These agroforests yield products for home consumption as well as for cash. The randomly mixed annual and perennial crops are of a variety of above-ground and below-ground architecture. Such a plant arrangement theoretically offers benefits of: protecting the soil from the eroding, compacting, and leaching effects of rain; high photosynthetic efficiency; preventing species-specific pests and diseases; and using the maximum surface and subsurface area. This describes a stable garden that is able to respond to environmental perturbations and degradation (Wiersum 1982). Such systems appear to be efficient in many ways. Interplanting can maximize the use of scarce garden land or land difficult to clear by utilizing the entire soil surface, subsoil, and vertical space above ground. Interplanting a variety of food crops is also a way of meeting the range of dietary needs. Because of the high diversity of species and their functions, such agroforests are argued to be both biologically and economically stable and sustainable (Michon et al 1986). Failure in a cash crop due to pests or a drop in international prices may be buffered by other crops (Wiersum 1982); also some component crop or tree species can be changed without affecting the overall ecological structure and productivity. Research on homegardens is most extensive in regions of high population density where homegardens have reached high levels of plant diversity, labour input, and percent of income derived from the garden (Padoch and de Jong 1991). More research is needed of this type of agroforestry in less densely populated regions. Throughout the economically poor tropical regions homegarden-type agroforestry is important as a source of subsistence and 7 cash resources, and a "repository and testing site for uncommon species and varieties of plants" (Padoch and de Jong 1991:166). 2.2.2 Polyculture and Polyvariety Polycultural agroecosystems have been seen as both an environmental and economic adaptation by indigenous cultures in tropical rainforests. However, Hames (1983) uses comparative data to suggest that polyculture is no more practiced than monoculture in tropical forest habitats. Hames argues that often there is more diversity of varieties of staple crops than diversity of species in a swidden; and polyvariety in itself can be adaptive and stable. Regardless, in acknowledging the great diversity of tropical environments, a model must be constructed representing the whole monoculture-polyvariety-polyculture spectrum. Whether or not ecological stability can be proven, it is evident that polycultural systems can provide economic security in a subsistence or semi-subsistence economy. Failure of one variety due to pests, drought, or frost can be buffered by other varieties. Polyvariety cultivation can extend the growing and harvest periods, or even increase production of a single harvest. At least there is some insurance that if some of the varieties produce poorly, others will do better, and the total harvest will be adequate. Clawson (1985) calls this a harvest security mechanism. 2.2.3 Polyphase agroforestry systems There are further issues complicating the characterization of traditional agroecosystems. Most researchers in the tropics carry preconceived ideas, and cultural biases into their observations of people, plants, and plant-human relationships (Alcorn 1981). A temperate zone forester or agriculturalist is not always able to see the patterns within the mess of plants said to be a garden. Many of the categorizations of individual plants and plant complexes must be redefined. 8 Gardens, homegardens, fallow gardens, secondary forest are often difficult for foresters to classify because they blend into each other in space and over time. Inputs and outputs may be invisible, and thus not quantifiable, and productivity difficult to measure. Typologies that place cultivated plants in preconceived spatial arrangements obstruct our understanding of plant-human relationships (Alcorn 1981). Many traditional agroforestry landscapes involve multiple phases through time or space. What may appear to be a simple swidden cultivation system may actually combine polycultural swiddens, interstitial treeplanting (planting trees along pathways, watercourses, and underutilized lands) (Raintree 1986), enriched fallows, and homegardens. Farmers can manipulate plot size, plot site location, density of crops, and crop diversity, within and between phases, to achieve production levels. 2.2.4 Characterizing horticultural techniques. Techniques and processes of cultivating individual plants in traditional agroecosystems may not fall into categories familiar to western agronomists or foresters. The process of domestication serves as an example. Plants are commonly called wild, domesticate, or weed by westerners. Wild plants grow in disorder, unmanipulated by man; domesticates are field-grown crops; and weeds are their undesirable companions (Alcorn 1981). But such distinctions are less clear in the tropics, especially in traditional agroecosystems. The Huastec Indians of Northeastern Mexico see spontaneous vegetation (weeds) as a resource, not as an impediment to agriculture (Alcorn 1981). The Huastec and the Maninjau of West Sumatra, Indonesia (Michon et al 1986) manage their forests by allowing elements of primary and secondary forest to coexist with introduced species. In the Maninjau multicrop system many species native to the original forest have been able to reproduce within the agroforest structure. Cultivation does not necessarily lead to domestication. A cultivated plant might not be a domesticate, and domesticates are not necessarily found in carefully cultivated plots. 9 There are many ways of manipulating plants, and only some of these cause the changes in genotype and phenotype associated with domestication. The domestication process depends on both human behaviour and on the plant's response to that behaviour (Alcorn 1981). There is a wide range of choices about treatment of vegetation. A plant can be slashed, neglected, spared, protected, transplanted, or planted. A particular plant will be treated in any or all of these ways depending on its position and usefulness at any one time. Decisions about manipulation are often based on seeking minimum labour input to render a plant useful (Alcorn 1981). When plants are manipulated en masse, a process of selection occurs that has different effects on different species, and local environments can be changed. Frequent slashing back could lead to the local extinction of a population that is unable to adapt. Other species that are adapted to disturbed conditions may proliferate. Thus human disturbance in rainforest areas influences the evolution of plant species by creating a variety of successional conditions. Of the different assemblages of techniques which could be termed agroforestry systems, there is variation in the degree to which the agroforesters modify their environment. Modification is best defined by the degree of "controlling and encouraging of certain species at the expense of others" (Ellen 1982: 125). Increasing the intensity of plant cultivation and plant selection generally creates a trend from generalized, species diverse systems, to specialized, low diversity ecosystems. But the trend is not always linear. Traditional agroforestry systems may involve some cultivated plants, some plants which are never manipulated, and some which are gathered but not modified. Species manipulation could actually increase diversity in some cases. In some marginal, low-density ecosystems, such as grasslands, the number of species may be increased by subsistence agriculture (Ellen 1982). Intensive, modern agrosilvipastoral systems may also increase species-diversity. 10 2.3 Traditional Exploitation of Melanesian Forest Ecosystems. Melanesia hosts a diversity of forest ecosystems in which the available resources differ, and thus a variety of traditional land use systems exist. For each system, the relative contribution of certain plant and animal species and the relative emphasis on different modes of resource utilization (hunting, annual gardening, tree farming, gathering) will differ. The range of ecosystems that are traditionally managed in coastal New Guinea include: coastal and estuarine mangrove and Nypa swamps, sago-palm fresh water swamp forests, dry eucalypt savanna and grassland, lowland foothill tropical forest, and temperate montane forest (Powell 1976). Human populations were probably initially centred in the coastal areas and on the rainforest-woodland and rainforest-grassland of the mountains; practicing a broad spectrum subsistence strategy (Powell 1982). With sea level rise and the consequent decrease in land and increase in population density, there may have been a gradual change of emphasis onto semi-cultivated plants such as sago, coconuts, bananas, breadfruit, and other fruit and nut crops together with yams and aroids; leading to the development of horticulture (Powell 1982). It is unknown whether plant domestication and agricultural systems developed with the help of outside influences. The available plant base would have certainly made independent development possible (Powell 1982). While recognizing the diversity, some generalizations can still be made about Melanesian subsistence strategies. Melanesians are primarily horticulturalists (Chowning 1977). Pure hunting-gathering societies were non-existent to very rare (Chowning 1977). Throughout most of Melanesia the food supply is based on a single root crop which differs depending on the locale. Taro is the most common where rainfall is heavy and steady. Yams are common in drier areas and on well-drained soils. Sago is gathered or cultivated in swampy areas. Sweet potatoes are dominant in the highlands (Barrau 1958). Secondary crops such as sugarcane, and a few greens, supplement the diet. Coconut and bananas are 11 important in the lowlands, as well as breadfruit and other fruits and nuts (Barrau 1958). Yams are the dominant staple in the D'Entrecasteaux islands. The lowland rainforests of New Guinea are especially rich in plant foods which alone could support small populations. Fruit and nut trees are plentiful, including genera such as Canarium. Pangium, Pometia. Spondias, and Terminalia. as well as edible ferns such as Dicranopteris, Gleichenia, and Pteridium (Powell 1982). Wild foods usually supplement cultivated staples and provide sustenance in the event of crop failure (Chowning 1977). Indeed the importance of both wild and cultivated fruit and nut species in Melanesian agriculture should not be overlooked. They are nutritionally superior to the root crops and palm products. They may have been the first group of plants to be tended and transplanted into groves and garden sites (Powell 1982). Barrau (1958) and Brookfield and Hart (1971) give a survey of the spectrum of existing Melanesian agricultural systems. Intensive cultivation systems have been observed throughout Melanesia involving ditching and composting sweet potatoes in the highlands; sophisticated irrigation techniques in New Guinea, New Hebrides and Fiji; and mounding in swamp forests of New Guinea. It is difficult to say how and why some of the more intensive systems were developed; perhaps in response to population increases and where arable land was limited. The pressure on resources may have led to the adoption or invention of adaptive technological advances facilitating denser occupation of the land (Brookfield and Hart 1977). Notable in many Papua New Guinea societies is what Brookfield (1972) termed social production; the production of food beyond what is needed for sustenance. Labour and land are used to produce crops and animals for cultural, economic, and political purposes^ . Allen (1988) would argue that social and political change and the demand for items of exchange in societies possessing strong egalitarian ethics, was the driving force behind technological development in horticulture, which in turn has created major and widespread environmental changes in the country well before the arrival of foreigners or foreign capital. 12 2.3.1 Agroforestry systems in Melanesia. Traditional agricultural systems of the Melanesian region have been typically characterized by a predominant taro and/or yam crop complex (Barrau 1965). However, ethnobotanical recordings in the southeastern Solomon Islands (Yen 1974), New Britain (Panoff 1972), the Bismarck Archipelago (Kirch 1989) and Fiji (Thaman 1988) have revealed that an arboricultural element in Melanesian subsistence is also widespread (Kirch 1989). Village agroforestry in these areas tends to centre on Pandanus sp., Arctocarpus sp., Canarium sp., Inocarpus sp., Spondias sp., Cocos nocifera and Barringtonia sp.(Kirch 1989). A distinct form of tree cultivation was dominant in Melanesian subsistence activities especially on smaller limestone islands and atolls which lack the land area or the edaphic conditions for intensive agricultural cropping. Such islands may have played a significant historic role in the development of agroforestry in Melanesia (Kirch 1989). Whatever the process of development of agroforestry, the result is at least 300 species of trees that have come to be known, used, and complexly combined into a variety of agroforests throughout Melanesia (Clarke 1990). The lowlands of the D'Entrecastreaux Islands are quite different biophysically from most Melanesian islands. So Fergusson Island may support types of agroecosystems representing a distinct divergence in the development of Melanesian agroforestry. 2.3.2 Human impacts on the Melanesian Environment Traditional resource uses have had different effects depending on the intensity of the activity and the local environment. In lowland rainforested areas, where people have practiced swidden cultivation with long bush fallows, a mosaic of vegetation types at all stages of succession is present. The dry woodland forests and savanah are often degraded to open grasslands as a result of burning for gardening or hunting. In the PNG highlands where 13 some of the most intensive and long-term land use occurs, grasslands are widespread and secondary forests only found on ridgetops and riparian strips (Powell 1976). Subsistence practices have different degrees of impact on the forest resource. Gathering forest plant products or tending plants in their natural environment, though a less intense activity than clearing and burning, can permanently alter plant communities. The mechanism is the removal of certain species, while selecting and encouraging others (Powell 1983). In insular island environments, sustainability of forestry and agriculture is especially challenged by intensive agricultural activities and resource utilization (Dahl 1984). Land use expansion on islands can easily produce changes in microclimates and soils (Tabora 1991) and destroy unique habitat types. Islands often support a concentrated combination of ecosystem types. Island biota are especially vulnerable to disappearance because the relatively small size of communities, habitat units, and populations can magnify the affects of human induced disturbance (Ingram 1989). The extremely high number of endangered species on South Pacific islands can be attributed to the above factors and to the introduction of predator and weedy species which displace island species (Dahl 1984). 14 Chapter 3. METHODOLOGY 3.1 Foundational Questions The objective of this study is to characterize the traditional agroforestry landscape at Nade, Fergusson Island, while emphasizing the collection of baseline information useful for future land-use planning. This study is by no means comprehensive enough to be used in planning. It is a basic inventory of the forest types and plant species that are modified by or significant in the lives of the people of Nade. Enough cultural understanding was sought to decipher the current patterns of land use and the forces of change. To structure the description of the agroforestry landscape the following foundational questions were used. i) Is the distribution of garden and forest types in the Nade landscape related to environmental factors such as geomorphology, soils, and topography? Site selection for cultivation will partly depend on the perceived fertility of sites within Nade. Some sites, because of topography or soil type, have limited cultivation potential. Biophysical conditions influence the process of succession, especially the species composition of different fallow areas. The shifting mosaic of vegetation types in the landscape will be a result of these factors. Classifiying and mapping the distribution of the successional phases at Nade was done using topographic maps, transects on the ground, and vegetation profiles. The phases were identified by vegetation profiles and a plant inventory. Field observations were corelated with data on geology, soils, and vegetation, obtained from the literature. ii) Can a spectrum of cultivated, semi-domesticated to wild landscape zones are managed by the people of Nade be defined by certain dominant plant species or plant combinations?. Previously, subsistence societies in rainforest have been coarsely categorized as hunting-gathering or swidden cultivators. Research of more modern agroecosystems recognizes only the fully cultivated zone. Many traditional systems are complex, involving a continuum of vegetation management phases, from fully cultivated, to semi-managed, to "wild " forest-like. Each phase is composed of wild, weedy, or cultivated 15 plants in different proportions. The phases also vary in the degree to which they are deflected from the process of natural succession. Information to test this hypothesis came from an ethnobotanical inventory, vegetation profiles of the various successional phases, and informal guided interviews with local gardeners. iii) Within environmental limitations, does the distribution of forest-garden types in the Nade landscape depend on a traditional system of kinship inheritance of land and/or plant tenure?. Decision-making regarding land-use is influenced primarily by the tenure system which could be seen to reflect the community's ecological relationship to the land. Tenure here refers to the set of customary rights held by an individual with respect to land and resources. Information on land tenure was obtained from informal interviews with villagers. 3.2 The methodology The methodology used in this study was derived from the interdisciplinary fields of development anthropology and ethnobiology. Development anthroplogy uses standard anthropological methods such as key-informant interviews, participant-observation, open-ended questionaires, and village census, to gather information useful to the planning and implementation of development programs (Green 1986). The investigation of the relationship of traditional cultures and their environment is the domain of ethnobiology, a hybrid field of anthropology and biology. Ethnobiologists weigh equally both the cognitive analyses of semantics and the gathering of basic biological and ecological data. From anthropology comes the recognition that "individuals from two different cultures inevitably think and speak with different cognitive 'realities'. For mutual interpretation to occur, sharing of 'realities' must also occur." (Posey 1990:48). To reach a 'sharing of realities' often takes more time than natural scientists or planners have. In a short-term study an anthropologist must rely heavily on what local experts tell him/her. The risk is that informants may have biases, blind spots, and limited perspectives. Data validity requires 16 sifting through often diverse and conflicting accounts and arriving at a coherent picture (Green 1986). If it is the detailed knowledge of biological and ecological knowledge that is of interest, then careful documentation, and checking and cross-checking to find anomalies and contradictions between informants is required (Posey 1990). For an open objective inquiry, care must be taken not to fall into archaic culture-biased dichotomies. For instance, that plants must be either domesticated or wild, indigenous peoples must be either hunter-gatherers or agriculturalists, and agriculture or forestry is considered mutually exclusive of agroforestry. The techniques applied in this study were partly derived from writings on rapid rural assessment (Rhoades 1987; Grandstaff and Grandstaff 1987; and Chambers 1987). Rapid rural assessment techniques have been developed to assess rural land-use situations when time, cost and available researchers are limited. It is commonly used to assess complex land use situations where integrated information of social, economic, and ecological domains is required. Some common techniques used in rapid rural assessment include: informal or guided interviews, using key indicators that integrate several variables, key informants, direct observation, and group interviews. All of these techniques are by nature logical and unstructured. Questioning and observation is done iteratively. Iterative process is highly suited to rapid exploratory learning as it allows changes in direction based on information obtained during the learning process itself, so that progressive understanding can occur (Grandstaff and Grandstaff 1987). The data collected using rapid rural assessment is qualitative, or in a simple quantitative form. Often the priority information is simply magnitude, and directions of change. To maximize the productivity of field research there is an emphasis on using existing available information as much as possible. Maps, aerial photographs, and satelite imagery can reveal alot about ecology, land use, and settlement patterns in remote and isolated communities. Admittedly, rapid rural assessment cannot replace a complete baseline 17 survey, but it can provide a first step for consideration of questions of evaluation of impacts, trends and causality (Chambers 1987). 3.2.1 Ethnobotanical inventory. An inventory of useful plant species was made to help to define the plant communities. Interviews were conducted about the use and management of the plants to determine to what degree the people of Nade are using and modifying those plant communities. Each plant species was listed by local names, corresponding botanical name when possible, life form, the habitat it occurs in, and its use by the people of Nade (Appendix B). Every plant with economic or cultural significance that was flowering or fruiting during the study period was collected. Ninety-eight plant species were collected and shipped to the National Herbarium in Lae for botanical identification. Seventeen of the medicinal plants recorded by Holdsworth (1974) for the D'Entrecasteaux Islands were confirmed to be in use at Nade. Other plants were identified to genus by using various references. And 136 species were identified only by local names. Location of collections is shown in Appendix D. Inevitably, when trying to corelate two radically different classification systems, in this case the scientific and the vernacular, there will be discrepancies. The investigator cannot assume the categories will match. One or more scientific species may be divided into two or three groups in the folk-classification: Varieties of one species may be lumped into another. The lack of fit does not necessarily indicate a poor observation by local people. It may just be a different logic. To systematically describe a folk classification is a project in itself. To relate that system to our scientific plant taxonomy, which itself is being continually revised, is beyond the scope of this study. The inconsistencies were reduced as much as possible by repeating questions and plant identifications with several informants. See the questionnaire in Appendix E for specific questions. Appendix A includes notes on the Nade classification system. 18 3.2.2 Maps and transects Classification of the successional phases at Nade was done using topographic maps, transects on the ground (Fig 2), and vegetation profiles (Appendix C). Mapping of the successional phases was conducted by making repeated walks along trails criss-crossing the study area in many directions (Map 7). Vegetation types were identified by eye and later verified by vegetation profiles. Maps were constructed by relating phase boundaries to key topographical features which were identified using compass and altimeter and a 1:25,000 topographic map (enlarged from the 1:100,000 base map). As the topography was very dissected (identifiable creeks 200-300 metres apart throughout the study area) the mapping could be done with reasonable accuracy by eye and with these simple instruments. The distribution of the 1989/90 gardens were mapped in the same way (Map 4 and 8). The spatial precision of the mapping for this study was limited to the accuracy of the 1975 base map. No maps were available at a scale larger than 1:100,000, nor were there adequate aerial photos or satelite images to help verify the mapping. 3.2.3 Vegetation Profiles Thirty-four vegetation profiles were completed, each covering an area of 50m ,^ in areas of different soils and topography, in fallow gardens and forest of different age classes. A twenty-five metre measuring tape was laid out through a representative section of a particular successional phase. All plants rooted within one metre either side of the tape were drawn to scale. The plants were identified in Nade language with the help of various informants. Notes were taken regarding location, soil type, topography, dominant species surrounding the profile, and general structure of the ecosystem. Representative profiles of the five major vegetation types are provided in Appendix C. These profiles revealled the relative frequency of occurrence of different plant species, and also different patterns and structure of the vegetation. A diversity of plant 19 communities could be classified from this information. But the vegetation patterns could not be simply related to natural processes. Each plant community identified would be manipulated by local people in different ways and to a different degree. Interviews about plant management techniques revealled the kinds of manipulation at work on the .structure of the various communities. A longer field study would be needed to determine the detailed effects of such manipulation over time. By selecting informants of different ages and sexes to assist with these profiles, the variation of plant knowledge in different sectors of the community and in different vegetation types, was revealled. 3.2.4 Garden diagrams Diagrams of the spatial arrangement of crops, in the 1989 gardens, were sketched to scale (Fig.3). During the research period, the various staple crops were continually being harvested and transplanted. Consequently, these sketches will only represent the gardens at one point in time. 3.2.5 Interview techniques Informal and guided interviews were used. For guided interviews no formal questionnaire was followed but a checklist of questions was used as a flexible guide (Appendix E). Informal interviews were more open, requiring mental and methodological flexibility. A successful informal survey does not proceed like the formal questionnaire survey where predetermined hypotheses are tested. Instead, new questions emerge as information is collected (Rhoades 1987). This is not to say the informal survey lacks logic, but that new information from interviews and observations are accommodated as the research evolves. Twelve men and 8 women participated in extensive guided and informal interviews on plant names, uses, and management. Numerous others provided insights through their 20 behaviour and through responses to casual questionning. Interviews were conducted in english with the 20 guides. Discussions with elders were conducted in Dobu language translated simultaneously into english by an english-speaking family member. Nade terminology for the plant world, land and resources was recorded to provide insight into how Nade people perceive and categorize their environment. 3.2.6 Land tenure The tenure system can influence the forest landscape via management decisions, involving for example, swidden site selection, removal of forest trees from secondary forest, or planting trees to claim land. Existing land disputes, and time constraints prevented the actual mapping of land claims. Instead, informants were questioned about the system of land and plant rights - how rights or ownership are determined, transfered, and maintained. The resulting data are normative statements about how the system is supposed to work according to the people of Nade. A more accurate definition of land tenure in Nade would require further mapping, extensive data on kinship relations, and a process of group meetings to clarify existing disputes. 3.2.7 Schedule of research activities I conducted the field research for this study in Nade over a period of 10 weeks, from September 15 - November 28, 1990. During my stay in Nade I lived in the home of the village schoolteachers^ , and their three children. They had the only house in Nade that was big enough for an extended visitor to stay, and the only house not made of bush materials; therefore the only house that the villagers were unashamed to offer .^ An advantage to such a living arrangement was that the teachers' house was centrally located adjacent to the schoolhouse, churches, and the mango tree under which community meetings are held. Though the schoolteachers are not from Nade, they are related to Nade families. Moreover 21 they both play central roles in the community by organizing school and church activities, by knowing everyone in the village, and by virtue of having fewer susu alliances, enemies, and secrecies^ . They could act as a neutral liaison and relatively unbiased translators. The research period coincided with the beginning of the dry season and the most labour-intensive phase of the agricultural cycle; the clearing, burning, digging and planting of the gardens. Thus, I had much opportunity to observe the swidden cultivation, but I had difficulty in finding local informants and guides with time to spend in the forest-fallow, and mature forest areas. Scheduling of my research activities was contingent on the availability of guides and informants. Most days, I hired a Nade villager to act as guide and informant to investigate the area most known to him or her. In any one day we measured gardens, collected plants, drew vegetation profiles, plotted vegetation types, and discussed land tenure. 22 Chapter 4. A BIOPHYSICAL DESCRIPTION OF FERGUSSON ISLAND 4.1 Location Fergusson Island is one of three large islands comprising the D'Entrecastreaux group, situated off the southeastern tip of the New Guinea mainland. Nade is a village of 35 hamlets situated on the southeast coast of Fergusson Island. Nade covers an area of about 50 square kilometres bordered by the Monayai river to the west, Selesilina River to the east, and the Edagwaba mountain range rising to the north (Map 1 and 2). 4.2 Geology and geomorphology Fergusson Island is 1340 km^ in area, with elevations to 1830 metres. The topography is varied. Some areas along the coast are steeply dissected with slopes reaching 50%. While other areas are flat hosting brackish mangrove swamps. The southeastern peninsula has little,water except for a geiser. There is debate over the geological history of the D'Entrecastreaux group. The islands may have been granite intrusions in line with the mountain axis of New Guinea and may have been connected to the New Guinea and the Australian mainland (Wood 1981). Alternatively, they may have been part of oceanic "amalgamation formed composite terranes" from basins far removed from the edge of the Australasian craton which then collided with the continental land mass (Pigram and Davies 1987). According to this theory the D'Entrecasteaux have had no land connection since the quaternerary. Dissected dome structures with granodiorite cores are evident on all the islands of the D'Entrecasteaux and there is debate over their formation. Oilier and Pain (1980) postulate that the domes originated by the pushing up of a granite pluton and foliation occurring during the rise. While, Davies and Ivies (1965) suggest that the domes are re-folded thrust faults. Fergusson Island has two such domes. Nade lies on the south edge of the Morima-Oitabu dome on the Oredi fault (Davies and Ivies 1965). Also notable is the volcanism on 23 Eastern Fergusson and Goodenough Islands. There are two volcanic cones on Fergusson, and two areas of geothermal activity. 4.3 Soils Soils in the D'Entrecastreaux group are predominantly inceptisols which occur up to 1500 metres. Soil information for Nade was obtained from (Bleeker 1983) (PNG Resouce Information System, PNGRIS, and field observations).The geologic transition west to east of the Nade ward results in an evident soil transition. There are four obvious soil units (Map 2). The hilly areas of western Nade are dominated by medium textured eutropepts, dystropepts, and troporthents on igneous and metamorphic rocks. These are moderately fertile soils suitable for a wide range of land use, including tree crops such as oil palm or rubber. They have a high percentage nitrogen and CEC, but with marked variation in individual nutrients. Available phosphorus and exchangeable potassium are moderate to low. The gently sloping areas of eastern Nade are coarse textured vitrandept soils which are typical lowland soils near active volcanoes. These are slightly weathered ash soils with a black topsoil, generally high in nitrogen, low in available phosphorous, and high in exchangeable potassium. These are fertile soils, suitable for tree crops, pastures, and arable crops. On the lowland coastal plain of eastern Nade are poorly drained alluvial fluvaquent soils, and well-drained coarse textured alluvial and beach soils called tropofluvents. Erosion potential of the Nade area is classified as "moderate" under the Universal Soil Loss equation (Bleeker 1983). 4.4 Climate The D'Entrecastreaux group is classified by the Koppen System as tropical rainy - no dry season, and by McAlpine and Keig (1983) as having a moderate seasonality and a rainfall regimen which is "continuously heavy". Mean temperatures at the Salamo airstrip, 10 km East of Nade, range from 30°C - 32°C to 24°C - 28°C in July. However, Nade ward appears to receive more rain than Salamo because of its location under the Edagwaba mountain range, while Salamo is situated on a flat volcanic fan. There are strong winds from the southwest from May through September. 4.5 Biogeography and ecology New Guinea lies east of the Wallace line, a transitional zone between the Asian and Australian flora and fauna. New Guinea has an Australian type fauna characterized by small, marsupials and lacking the large Asiatic mammals. The botanical boundary is not as clear as the faunal one, placing the flora in the Indo-Malesian realm (White and Connell 1983). More than 90% of New Guinea is forested, mostly with lowland rainforest. Various vegetation zones are evident, and they are more closely related to altitude than rainfall (Paijmans 1976). The high diversity in relief and habitats in New Guinea, encourages speciation. This may account for the high number of animal and bird species in New Guinea, relative to Australia. Climatic fluctuations have influenced the vegetation of Southeast Papua. Much of the humid rainforest of today was dry forest and savanah 30,000 - 12,000 years ago. Lowland rainforest became the dominant vegetation type only 8,500 to 5,000 years ago. At this time evidence appeared of human impacts on vegetation related to burning (van Balgooey 1976). Detailed ecological studies relevant to the D'Entrecastreaux Islands are lacking. Brass (1959) documents a 1956-57 expedition made to the D'Entrecasteaux islands during which some botanical collections were made. Paijman's (1976) classification of the major vegetation types on Fergusson Island includes: cloud forest (with Castanopsis sp.), lower montane forest (dominants of the common mixed types have not been determined), small crowned lowland hill forest (with pure stands of Casuarina sp., Castanopsis sp., Hopea sp.), small crowned forest on plains and fans (poorly identified dipterocarps), and sclerophyll scrub (dry areas dominated by eucalypts). Ingram (1989) provides an expansion of Paijman's physiognomic types and also lists some vulnerable elements of local biological diversity. 25 Diamond (1972) has studied the biogeography of Fergusson Island with reference to the avifauna. Lecroy et al. (1980) studiedthe Goldie's Bird of Paradise and its relationship to the island people. They also published a list of bird observations on Normanby and Fergusson (Lecroy et al. 1983). 26 Chapter 5. T H E HISTORICAL AND CULTURAL CONTEXT 5.1 The Massim The people of the D'Entrecasteaux fall into a broad cultural category first described by Seligman (1910), and refered to as the Massim. They are island traders who live on the steep volcanic or low coral islands of Southeastern Papua New Guinea. The material culture (Newton 1975; Beran 1980) and social customs of the Massim are distinct from those of neighbouring areas. The Massim are among the minority of New Guineans who speak an Austronesian language and have a matrilineal social structure (an exception are the Goodenough Island people who are patrilineal). A further distinction can be made between the northern and southern Massim. For the Trobriand and Marshall Bennett Islanders in the north, social rank and leadership are mostly hereditary, and village alliances are partly created through polygyny and yam contributions. For the southern Massim of the D'Entrecasteaux influence is earned by personal accomplishment, and polygyny is rare (Beran 1988). The island communities of the Massim had local and specialized manafacture of various goods, and they participated in utilitarian trade. Some of the trading still takes place on the voyages made for the complex and extensive ceremonial exchange of shell valuables called kulcfi. Not much is known about the history of the Massim before Western contact. Good archeological records are lacking. There are anthropomorphic rock engravings in the D'Entrecasteaux, but nothing is known about their age or origin. Pottery sherds found in deserted village sites suggest some of the islands have been inhabited for up to two thousand years and that pottery trade and design changes occurred during this period (Irwin 1983). 5.2 European influence The first European contact in the D'Entrecasteaux was in 1793. Western commercial activities and Christian mission stations were established throughout the Massim area in the 28 late nineteenth century. The area became part of the British protectorate in 1884. The first mission in the D'Entrecasteaux was Methodist and established at Dobu Island in 1891. The hospital at the Salamo Methodist mission station on Fergusson Island opened in 1925. The Roman Catholic established a mission at Budoya on Fergusson Island in the 1950's. The Melanesian people of Fergusson were living mostly in the interior until the early 1900's. The Galia and Ebadidi tribes in the interior were not "pacified" until 1920. Young (1971) and Jenness and Ballantyne (1920) characterized the seventy years of European influence on indigenous settlement patterns and local groups of Fergusson and Goodenough Islands in terms of three broad trends. There was a general shift of populations from isolated upland settlements to more easily accessible lowland sites; a tendency for dispersed hamlets to become nucleated; and an increase in the overall size of communities. As Dobu Island was the first to be missionized in the area, Dobu became the official school and church language. Dobuans acted as missionaries on Fergusson Island and they also supplied the interpreters and clerks who accompanied government officers. The result was the rapid substitution of the Dobuan language for local Fergusson languages (Chowning 1969). Dobu is now the lingua franca for most of Fergusson Island (Lithgow and Staalsen 1965), though different villages retain a few words of their original language. In Nade, Dobu is the commonly spoken language. The Nade language is known only by a few people over the age of about 60. English is taught in the village school up to grade six. Since the beginning of European contact, acculturation of different groups within the Massim has been facilitated by greater mobility and the focal location of towns, schools, and missions where people of different cultures can meet (Chowning 1969). D'Entrecasteaux men have migrated to jobs on plantations and mines on the mainland and other islands, returning with cash, goods, and sometimes customs of other cultures. Inter-island marriage has become increasingly common. Patrols, initiated by the Australian administration starting in the 1940's, began a network of communication. Patrol officers brought news from afar, 29 and monitored the state of health and production of remote villages, sometimes intervening in cases of drought and famine (Young 1971). Missionization and acculturation between D'Entrecasteaux communities have blended more than just language and customs. European missionaries introduced some food plants such as citrus, english potatoes, squash, and corn, all of which are seen very occasionally in local gardens. At the high school in Salamo, students are introduced to gardening techniques and given seeds for some non-traditional foods. But even when they take them back to their village they aren't used. Missionaries also introduced ornamental flowers such as zinnias and marigolds. In the early 1900's Fijians were recruited as village pastors and they brought food plants native to Fiji. They also introduced species of Pandanus used for decorating woven mats, with the accompanying weaving techniques. There is no large-scale cash crop production in the D'Entrecasteaux Islands. Copra was produced (at its peak in the 1970's) on an individual and family scale, and sold at low prices to the local European traders or personally taken by boat to the copra marketing board in Samarai (Young 1971). Other cash crops such as coffee, cocoa, chillies, peanuts and rice, have been experimentally grown in various locations under the direction of administration agricultural officers but none have proved successful (Young 1971)^ . Islanders' cash income comes mainly from migrant labour. Timber resources are considerable and have just begun to be exploited. A few small saw-mill ventures by local entrepeneurs have failed through lack of capital and technical knowledge. The commercial logging of Ulabo Timber^ operated on the east coast of Fergusson Island from 1987-90. 5.3 The Establishment of Nade. Nade began to be settled in the early 1900's soon after the Methodist mission was established at Dobu. Oral history suggests that the two prominent lineages of Nade 30 immigrated from the interior near Salakahedi and from the east at Morima (Duduwega 1975). Other lineages immigrated later from Normanby and Goodenough islands. People traditionally have lived in hamlets of 2-6 households. Originally these hamlets were spread out across the hillside, usually built on knolls and flats of ridgecrests with a good view of the ocean and approaching enemies. Informants claim that some clans had the special function of patrolling the hillside at approximately 400m elevation, watching for intruders coming from the interior. Originally, Nade referred to the cluster of hamlets at the beach, the western half of the present Nade ward. The interior hamlets in the Selesilina valley, in the eastern half of Nade ward, were collectively called the village of Ogauga. The villages were effectively divided by the Edagwaba ridge. Since missionization and the Australian administration some upland hamlets moved to the coast, and there has been more intermarriage. Nade has become a string of 23 hamlets along the beach, and 11 further inland on the coastal plain (Map 3) with a population of approximately 390. Nade and Ogauga are now administered together as Nade ward under one local councillor. However, the old division remains significant for the Nade people and community work is often divided between the two sides. People may have moved to the coast to be closer to mission services. The two churches, United Church and Roman Catholic, and an elementary school are situated at the Nade mission station near the beach in the centre of Nade ward. The once relatively autonomous and even adversarial hamlets now function more collectively. Larger social activities take place as accessibility and communication are facilitated by proximity, and the churches provide a focal point. Though only a few individuals from Nade have been to Port Moresby, many have been to the provincial capital of Alotau, and they have regular contact with other centres in the D'Entrecasteaux. People of Nade make regular trips out to Salamo, the United Church mission which administers an airstrip, post office, trade store, medical clinic, and high school. Salamo is a two hour walk along the beach from Nade. An inland trail is overgrown 31 and unused. Four Nade men each own a small dingy and outboard engine which they hire out for transport to Salamo, the Catholic mission at Budoya, or the district administrative centre at Esa'ala on Normanby Island, all about 0.5 hours by boat from Nade. A few Nade villagers can afford to pay with cash from relatives in Moresby. 5.4 Nade social organization No anthropological research has been done at Nade. However, some inferences might be made from other writings on the Massim people - in particular, Chownings' (1962,1969,1983,1989) work at Molima (15km west of Nade), and Fortune (1932) at Dobu. Chowning (1969) observed increasing acculturation of Molima with the Dobuan culture. Due to the proximity of Nade to Dobu, the process of acculturation could also be expected. Indeed, Dobu is the language commonly spoken at Nade, and there has been intermarriages with Dobuans. As are most of the Massim societies, Dobu and Nade are matrilineal. Social organization is based on a susu - the unit of a woman, her mother, her mothers' sisters and brothers, her sisters and brothers, her children, and her brothers' and sisters' children. Each susu, or matrilineage, claims descent from a common ancestress, through females only. A susu only knows its own genealogy back for approximately four generations and the common ancestress is a bird not a human being. Susus sharing the same bird ancestress belong to the same totem (Fortune 1932). Each totem, named after a bird, knows its legend and each susu knows its ancestry. Each susu has its own hamlet. Marriage is hamlet and susu exogamous. Residence is often, but not strictly matrilocal. Garden land, village land, personal name, status, village palms, fruit trees, and seed yams are all passed through the susu, mother to her children, father to his sister's children. It is in the institution of marriage we begin to see the overlap of the European patrilineal system of inheritance on the traditional matrilineal system. Some Nade people have Christian church sanctioned marriages in which the woman takes her 32 husband's name. In Nade, the couple each live as members of their respective matrilineal susus, but in any contact outside of Nade with missions, with government administration designed by Australia, or with the market economy in general, the woman takes her husbands name and status. Though matrilineal inheritance is still the rule in Nade, this increasing contact with patrilineal systems may have future implications for how land is transfered. Forces of change in the inheritance system will be expanded on in Chapter 8 5.5 Exchange - The Massim perspective on land and food. It is a truism that Melanesian peoples in general value food in ways which transcend its intrinsic value for them as a necessity of life. Its valuation is such that it appears to be used everywhere to create, maintain and manipulate social relationships (Young 1971:146) Kahn (1986) collected ecological data at Wamira, mainland Milne Bay, to confirm that the Wamiran preoccupation with food and hunger has little basis in ecological fact. Rather food is an important mode of symbolic expression...Wamirans use food as a vehicle with which to communicate gender qualities, control ambivalent relations between men and women, and manipulate political rivalries among men (Kahn 1986:1). In Nade too, people frequendy expressed concern that they may not have enough food, that the yam gardens are all they have. Taking into account the variety of foods available in the different forest/garden types (see chapters 6 and 7) famine should be rare. Historically, famine was recorded on nearby Goodenough Island and northern Fergusson Island in 1899-1901, 1911-12, 1946-47 and in 1957-58 (Jenness and Ballantyne 1920; Young 1971). It is unknown how Nade was effected. In Melanesian societies food represents more than caloric intake. It is essential to the reciprocal exchange system that is the foundation of social organization. Food production usually emphasizes one symbolically and culturally important crop supplemented by other more utilitarian food plants. Such a production system may not be efficient when yields as defined by agronomists, are measured. A cost-benefit analysis, even when loosely applied to the subsistence economy, may reveal glaring inefficiencies because economists cannot account for the cultural values of food. To an outsider at Nade, an inordinate amount of time, labour and resources is expended on the preparation of feasts which are usually associated with memorials to the dead .^ For most Massim societies it is yams that are central to life. Yams are variously the focus of magic ritual, central in myth, personified with gender, essential to commemorate the dead; and the basis of reciprocal exchange systems (Fortune 1932; Malinowski 1935; Young 1971; Weiner 1976). 34 Chapter 6. PLANT INVENTORY 6.1 The importance of plants in the Nade subsistence economy The plant world is extremely important in the lives of the people of Nade. Plant products are the foundation of the subsistence and cash economy. Local plants play a key role in cultural expression through their use in traditional rituals and dance, and their importance in myths and legends^. Plants are symbolic tokens in the maintenance of social relations through their use in sorcery and witchcraft, traditional medicine, and food exchanges. A total of 290 undomesticated plant species of economic or cultural significance were identified and are listed in Table 1. Most of these plants are found in the fallow gardens, and lowland forest. Many are selectively left or transplanted into the mixed sago stands, and the village agroforests (see Chapter 7). Species in the montane forest are less well known or used. The majority of the plant species listed are gathered or selectively maintained and not formally cultivated. The eleven domesticated crops cultivated in swidden gardens are listed below. This list includes a group of 7 staple food crops locally called tabuwala. These are few in species number, though they represent the plants of basic sustenance. Yams are the most important plant cultivated in the gardens in terms of quantity and cultural significance. Nade people consider yams the most important of all the foods. Indeed, yams are placed in a category to themselves while all other plants are grouped as asowa. I was not in Nade long enough to observe the full symbolic significance of yams in social exchange. However, it was apparent that yams are the special food to serve to guests at funeral and ceremonial feasts, and they are the measure of food security, of gardening ability, and of social integration. Though the relative use of the forest plants was not quantified, the number of species listed does demonstrate the importance of uncultivated plants in the subsistence economy at 36 Nade. Quantifying the value of many of these plants, in terms of frequency of use or product equivalents in the cash economy, is problematic. Moreover, calculating the number of species in use, and the extent that they are exploited is difficult because the knowledge of plants used for healing, and for sorcery and witchcraft, is secret within susus and often gender-specific. Evaluating the importance of medicinal plants in Nade society cannot be reduced to an equation of the replacement cost of western drugs. These plants individually or as a group represent a cultural importance that is impossible to enumerate. The use of plants for construction materials, foods, and gardening equipment may be easier to quantify because they have equivalents in the trade store on the island, but few villagers have cash to purchase them. Some of the plants listed are no longer in common use, having been replaced by cash products. Some uses of multi-use species are obsolete. Fishing line, clothing, and flashlights are examples of goods that are no longer obtained from the plant world. Inexpensive and functionally superior equivalents can be found at the mission trade store. Construction materials are never bought by villagers, though axes and machetes used to obtain them are. Food is not bought though it is occasionally supplemented by cash products, for example, canned mackerel and rice. Throughout Melanesia, traditional cultures are increasingly integrated in the cash economy, resulting in decreasing dependence on the resources in their environment. With a dependence on Western clothing, materials, and fibers, native uses and knowledge of traditionally used plants also declines. In many cases, the native and aboriginally introduced plant names have been forgotten. I heard repeated laments by the elders that younger people no longer know the important plants by name. I probably left Nade with a better knowledge of the useful plant species than anyone my age. Some of the more important species will be discussed here. Plant knowledge is generally divided between the sexes and between clans. Men are most knowledgeable about plants used for construction materials, and plants such as poisons 37 used in fishing and hunting. Women are most knowledgeable about food plants such as ferns and edible leaves and roots, plants used for household utensils, and dyes and perfumes. The plants can be divided into categories based on local use. Indeed, if an informant didn't know the local name for a particular plant he or she might give it the general name for its' use - creating some confusion in documentation. These categories are not exclusive; multi-use species are important in a few categories. From one tree the wood may be used for construction, leaves for medicine and fruits for food. 6.2 Plants cultivated in the swiddens There are three species of yams planted at Nade, and many varieties. CSiakutu Dioscorea escuelenta), and kwatea (Dioscorea ajata) (Purseglove 1972) species are planted in the greatest quantity, and hold the most symbolic significance. [Noma Dioscorea) spp. is of minor importance. Taro (Colocasia escuelenta). various types of banana (Musa spp.) and cassava (Manihot escuelenta) (Purseglove 1972; Williams 1979) are staple foods that are only cultivated in swiddens, but culturally they are perceived as supplementary to yams. Papaya (Carica papaya), sweet potatoes (Ipomaea batata), and pineapple (Ananas comosus) (French 1986) are more recent introductions. Sugar cane (Saccharum officianum) (French 1986), papaya, and pineapple are planted in the swiddens for snack foods to be eaten during a working day. Aibika (Hibiscus manihot) is also introduced and is the only cultivated green vegetable. 6.3 Multi-use plants Palms such as sago (Metroxylon spp.), coconut (Cocos nucifera) and black palm all have multiple uses. The leaves are woven together for mats, sails and roofing; or finely stripped for grass skirts; stems may be used for construction, root fibers are used for fishing 38 line, fruits and young shoots may be eaten, and parts may be used for medicine or magic. Sago provides an abundant source of food starch. At least eleven species of the Pandanus genus occur in various habitats in Nade (Table 1). Some of the multiple uses of the genus include: weaving the fronds into sails, sleeping mats, small shelters, and blinds. The fiber from aerial roots is used for making fishing line and net, tying roof thatching and clothing. A common beach Pandanus species has an edible fruit. 6.4 Food plants The main food plants are root crops cultivated in swidden gardens. These are supplemented by fruit trees, both cultivated; papaya (Carica papaya), banana (Musa sp.) and wild mwagolu (Szygium mallacense). piauwa (Pangium edule), and damaia (Ficus copjosa), and daewa (Arctocarpus spp.). Three types of mango (Mangifera spp.) are recognized; one is introduced and called the white man's mango, another is the native mango which is transplanted near housesites and gardens, and a wild mango is found occassionally in mature fallow gardens. Edible greens, known as pai, are gathered from the forest and represent an important supplement to the diet year round. Many of these greens are ferns such as beluwai (Sphaerostephanos ajatejlus), and ubuga (Callipteris prolifera'). While some tree and shrub leaves are also eaten such as kegalu (Gnetum gnemon), and kalaneania (Ficus sp.). Sago is important as a food supplement between harvests of yams or when the yam harvest is poor. A starch is obtained from the sago trunk, and prepared for eating by roasting or boiling. Buyado is a term for a category of plants used for seasoning in clay pot cooking, for example Mogomogo (Athyrium silvaticunf). and wasewasegala (Mapania macrocephala). 39 Large leaved plants are used for wrapping food to cook over open fire or hot stones. For example, lucoluco (Heliconia bihai), piesi ("Cordyline fruticans-) and a few Pandanus species, and banana (Table 1). 6.5 Construction plants House construction materials are usually gathered in the modified small-crowned hill forest, lowland forest, or fallow mosaic. Many of those used for posts and beams are also commercial forest species such as Pterocarpus indicus. Alstonia scholaris. Intsia spp., Terminalia spp., and Dillenia castanaefolia (Eddowes 1977), and are found in mature fallow, or modified forest. Others, such as Homolanthus novoguiniensis. Commersonia bartramia. and Macaranga densifolia. are found in earlier successional phases. Flooring is commonly made from stems of kakolu palm, black palm, or bamboo. All the roofing and walls of the residential houses, and often the yam houses too, are made from thatching made of sago leaves. If sago is scarce then kakolu or nipa palm is used. Axehandles, knifehandles, and spear shafts are carved from the hard inner core of the black palm and other hardwood timbers such as ilimo, Terminalia spp. Canoes are commonly carved from Intsia sp., Pterocarpus indicus, Glochidion novoguiniensis. or Alstonia scholaris. 6.6 Fiber plants The fiber plants, mainly coconut, sago, and various Pandanus species, are used for weaving mats, sails, and baskets. Coconut or Pandanus leaves are stripped and dried to make grass skirts and men's loin cloths. This traditional clothing is only worn daily by elders now, but remains important in dancing and cultural activities for everyone. The root fibers of these plants are used to make fishing line, and fishing nets. Various vines, collectively called waige, are used for tying flooring, posts and beams, and roofing in place. Some fibre craft techniques were introduced. Fijian pastors brought a species of Pandanus and a technique for dying and weaving patterns into the mats. 40 6.7 Medicinal and magic plants Many of the medicinal plants recorded in this study are common early successional forest species, or weedy in old swiddens. Some medicinal herbs may be encouraged in certain sites in the garden, or near housesites. Sixty-seven different medicinal plants were recorded. Plants are used as contraceptives, abortificants, and to aid in birthing. They are also used to treat colds, pneumonia, malaria, typhoid, tuberculosis, and dysentry. Most medicines are derived from the roots, leaves, or bark of forest plants. They are administered in the form of teas to drink, or to wash in, or the plant material may be crushed and applied externally. In this study, plants are characterized as medicinal when the informants stated that the plant is used to help such disease symptoms as would be described by western medicine. In fact, for the people of Nade the category of medicinal plants is not clearly distinguished from that of plants used for sorcery and witchcraft. They explain most disease or injuries as misfortunes caused by deliberate acts of sorcery (done by a man), witchcraft (done by a woman), or malevolent forest spirits. Treatment often calls for a traditional healer to divine the sorcerer or witch or spirit responsible for the ailment, and then to use plants and words to effect a cure. The most powerful magic/medicinal remedies, for the more serious illnesses, consist of a combination of specific plants administered with words or spells, by a specialist healer. For each of the common ailments, such as colds, cuts, and malaria, one of many plants are used as remedies and administered without words or spells. Medicinal plants for common ailments are general knowledge, though each susu also has its own. Fortune (1932) observed that introduced diseases such as measles, tuberculosis, influenza, and dysentry are often recognized by Dobu people as being of introduced origin. This is also true in Nade, though some introduced diseases have now been included in repertoires of magic. In Nade, people showed me some plants that they use to treat these ailments. At the same time, local people readily accept western medicines, for treatment of 41 the introduced diseases or for the common boils, infected cuts, and malaria symptoms. They frequent the village aid post or the mission hospital three hours walk away seeking cures for a variety of ailments. It seems that in the minds of the people of Nade, western drugs and medicinal plants have a similar function, the relief of symptoms. Indeed, western medicine can be very powerful for symptom relief. However, there is a sense that in some cases western medicine cannot go far enough to cure the cause of disease. When all disease was believed to be the result of sorcery and witchcraft, it was tensions in social relations that caused a sorcerer or witch to inflict disease**. Now, when an individual is hurt or sick they are still uncertain of the cause. They may try western medicine but if there isn't a quick cure, then they would have to assume they are the target of a sorcerer's or witch's attack. Magic is most used in connection with matters of importance and uncertainty, such as weather, gardening, hunting and fishing, sexual attraction, trade, warfare, sickness and death (Chowning 1977). It may be used when making canoes or weapons, but not usually for everyday tools. Men and women all admit to knowing some plants for white magic, that is for garden magic, love magic, and magic as protection from forest spirits. This knowledge would be passed mother to daughter, father to his sister's son. Knowledge of the most powerful plants in sorcery and witchcraft, for weather control, and for creating or healing serious illness is the domain of certain individuals passed discriminately through the susu. Young people speak of having to be worthy of the knowledge. They must take an interest in and care for their parents or uncles and aunts in order to earn it. Knowledge of black magic, magic to inflict disease or kill is never admitted* .^ Examples of plants found in the Nade landscape are as follows. The fern ally Ataito'ala (Selaginella velutina) is in constant use as protection from forest spirits. Women place it in their own and their children's hair when they walk in the gardens and forest. Kaladimumu (Ficus wassa) serves the same purpose. Various ferns and herbs are used as 42 love magic. Other plants are used to improve the voice. Katakdta'iuna, (Ficus in sculp ta") is used to smear on the lips to facilitate negotiations during the Kula trade. The category of plants believed to enhance yam growth and protect the yams from theft is called Swalolo. These plants are crushed into cold water and the mixture is squeezed over the yam seeds immediately before planting. The process itself is not secret, completed at the time of planting while the group of women are gathered to plant. However the plants themselves are usually secrets held by each susu^. 6.8 Ornamental Plants Plant products are used in various ways to decorate the person or the home. This category includes plants used as dyes for the fiber crafts such as Pandanus spp. and Aigwama (Acalypha hispida). and kwanakwanamatadi (Breynia racemosa'). Perfumes are made with fragrant plants such as Anelewa (Cananga odoratal and coconut oil. The leaves and flowers of other plants such as kemwata (Euodia hortensis) are used to decorate the body during dances. (Table 1). 6.9 Plants of ecological significance Several different plant species are recognized as season and weather indicators and are used to indicate when to undertake the various gardening or hunting activities. It is said that when the flowers of Maluwana (Toona sp.) drop then there will be the rains preceding the dry season. When the leaves turn red then the winds are calmer and it is time to fish and time to clear gardens. Mwadawa (Pterocarpus indicus) marks the start of the dry season. When the leaves fall it is time to burn. When the flowers come out it is time for collecting crab. When the flowers are washed off by the rains, and the new leaves come out then it is time to have the yam seeds planted. When Frangipan flowers it is time to dig the garden. When Dagwenau and bidubidu (Barringtonia spp.) flower it also marks the 43 beginning of the dry season. When their leaves turn colour it is the time to search for wildfowl eggs. Plants of more direct ecological significance are those that are used by the local community, and which also play a key role in the structure and function of the ecosystem, for instance, plants that are important regulators of energy and nutrient flows in the agro-ecosystem. Mwadawa (Pterocarpus indicus") is commonly harvested for construction of houses and canoes. While living it is used for fencing the yam gardens and as yam supports. As a fast - growing, coppicing species the roots of Pterocarpus indicus may be important in preventing soil erosion in the steeper yam gardens. As a nitrogen-fixing species in the Leguminaceae family, Pterocarpus indicus undoubtedly plays an important role in enhancing soil fertility in both the new and fallow gardens. 44 Siakutu (yam) kwatea (yam) potatoes cassava taro banana sago papaya breadfruit pineapple sugarcane citrus chestnuts ocarynuts mango bush-yams aibika (greens) ficus (fruit) ubuga (fern) piauwa (fruit) salewa (fruit) totosi (fruits) beluwai (fern) kegalu (greens) Fig. 1 Seasonal Jan ***: :**** | c * * * * * * * * 3 | * * * * : ! * * * * ***::**** ********** ***** **** ***::**** ***::**** ***::**** *** *** ***::**** ***:<**** * Feb | e * * * * * * * * : i * * * * : ( * * * * | e * * * If********* * * * * * * * * * * * * * f***-.t**** * * * * *************** ****$**** ****}*****\***** ********** ***: ***: ***::****)*: **** * * * * * It * * * : e * * * * |c * * * : c * * * * * * * * * Mar Apr availability of foods May * * * * 4 * * * * l * * * * j * * * * l 4 c * * * 4 * * * * * * * * * * * *********i****i****k****)***** ****** *********=********* * * * . : * * * * * * * * * ****i:***** * * * * ****:'(***: {***! : * * * * * * * * s > * * * * «| s(s s(s sfc >fc Jun !<***<*** !<***<***: Jul Aug ************ Sep **** * * * * 3 j ***** { e * * * | c * * * : < * * * * * * * * : < * * * * * i C * * C «fC * * * * * * * * * ****: * * * * c * * * * * Oct |C * * * >^C <fl i e * * * * * ***** ***** ***** * sfc s|c *fc Nov Dec ***** ***** *********** *********** ********** cfC «fC * ***** ****** ***** ****** ****** *****: |c * * * sfc * ****** *****: Chapter 7 CLASSIFICATION. DISTRIBUTION. AND MANAGEMENT OF  SUCCESSIONAL PHASES OF THE NADE CULTURAL LANDSCAPE 7.1 Classification of successional phases in the Nade cultural landscape. In Nade, it is immediately apparent on walking in "the bush" or mitawa (the Nade term for all forest land that is not village and not currendy a garden) that one is moving through a continuously changing mosaic of plant communities. Most patches have begun their present development after human induced disturbance, such as swidden clearing. One can observe common species persisting in various habitats, and at varying stages of succession, and those requiring a more specific habitat or successional phase. The people of Nade acknowledge the mosaic and have terminology to describe it. They travel through mitawa daily, to and from their gardens, gathering firewood, medicines, and foods. Within their own susu lands they know the history of the mosaic; when each patch was bagula or new swidden, and when it transformed to various stages of fallow (yakwala, yakwayakwala, and waudala). They have observed the relative rate of growth of different sites, they know where to look for a particular plant species, and they know aspects of the behaviour and habitats of the local wildlife. The following types of ecosystems and successional phases in the Nade cultural landscape were classified according to the identification of plant components, and the observation of management activities of Nade gardeners. Their distribution is shown in Fig.2, Maps 4,5, and 9. I. New and one year old swiddens, cleared 1989 and 1990 (bagula) II. Fallow sites, cleared 3 to 20 years ago (yakwala, yakwayakwala, waudala) III. Hamlet agroforests IV. Coconut and betelnut groves V. Mixed sago palm groves ('o'ai) VI. Slightly modified small-crowned hill forest (uwama, mitawa) VII. Slightly modified lowland forest (uwama, mitawa) 46 VILT. Primary montane forest (uwama or mitawa'ai) IX. Nipa palm swamp X. Mangrove swamp XI. Grassland XII. Beach XIIJ. Housesites XIV. Cocoa and Coffee XV. Riparian I. Yam garden management and design Nade swiddens are primarily yam gardens. Yam gardening involves the most important sequence of activities in Nade community life - a reflection of the high value placed on yams. The yam gardening cycle dominates the Nade calendar. It is a yearly process timed by one wet, one dry season of clearing, burning, grid-laying, planting, weeding, harvesting, and storing. In September, when the Pterocarpus spp. drops its flowers indicating the beginning of the relatively dry season, the men clear the new yam gardens. First, they cut away the understory of shrubs and vines using machetes. Then they fell the bigger trees and lop the branches off using axes. The site is left to dry for 2-3 weeks and then burnt. A group of 4-6 men can clear a garden in one day or an individual will take about one week. Each man and woman has their own garden of 0.04 ha to 0.4 ha per year, depending on the number of dependents. If the swidden is cleared but too much rain prevents drying then the men must drag the debris off in order to plant crops. In this case, according to informants, more weeding will be required than on a burnt site. Gardeners have no articulated conception of the nutrient contribution of ash to the soil. Rather, they see burning as necessary as an efficient way of clearing the debris and to prevent excessive weeds, thereby reducing labour. 47 Hamlet with houseslte vegetation Beach Hamlet agroforest Sago grove Modified lowland forest Fallow Betelnut grove Swidden Small-crowned hill forest Montane forest XIII XII III V VII II rv I VI VIII I(cleared 1990) Kcleared 1989) II (cleared 1988) 650 •600 550 500 450 400 350 [300 i-250 200 150 100 50 Om 2.0 km North • Fig.2 Schematic cross-section of transect between Niania and Waipe Creeks 0 0 (located on map 7). g Map 5. Distribution of successional phases of the Nade agroforestry landscape. Swiddens are completely cleared. Gardeners believe that yams do not tolerate shade, and the litter from the trees is "messy". Complete clearing helps reduce the seed source of weeds, though this was not directly articulated by informants. The men cut all the stems that are greater than 5 cm diameter at an average height of 1.5 m, the rest they cut flush with the ground surface. The standing stumps will vegetatively propogate and will be used to train the yam vines. Common species left as stumps are Pterocarpus indicus. Commersonia  bartramia. Homolanthus novoguiniensis. and Hibiscus tiliaceae. The women gather and pile any small debris that is not burnt to burn again. They also pile and burn the debris after a harvest. They believe that constant smoke in the garden will drive away insect and bird pests. After burning, men or women lay the charred logs in a grid pattern 2m apart. On slopes, they place small stakes or rocks below to support the horizontal logs. This practice has been shown to reduce soil erosion as demonstrated in Thailand by Sabhrasi (1978). The gardeners of Nade also articulate this purpose. Men or women lay a grid pattern on flat gardens as well (Fig.3). The grid divides the garden for the purpose of organizing the crops and the work within an individual's garden, or to divide the swidden between individuals. Every man and woman has a garden of their own in which they plant their own yam seeds. Siblings or cousins or a husband and wife may garden in the same swidden, so their gardens and seeds will be separated by logs. Men build the fences to keep out wild and domestic pigs. To save labour fences may be built communally around two seperately owned gardens. While clearing the garden the men strategically leave stumps of, most commonly, Pterocarpus indicus and Barringtonia  papuana. where the fence will be built. They supplement the rooted stumps with vegetative cuttings of the same species to form two lines of posts 15cm apart, then tie cross beams between them using vine stems. The living, rooted posts make a stronger fence than digging in sticks and logs. 51 «- 57m-» -f • O ->—h -t—+• -t—r - i — h 0 0 0 a • • • • • • • o 0 0 o • • • • • • 0 • • 0 o 0 B 0 0 0 0 0 0 0 0 o 0 1 1 — ( — • — J - H — I — M 1-Garden division t. Yams (D.Alata) • Yams (D.Escuelenta) • Taro V Cassava • O • 0 0 0 0 0 0 0 V 0 0 0 • • o 0 o 0 0 0 0 0 0 0 0 Log grid — Sweet potato Banana Papaya 0 T A Fencing i • • • Pineapple A Sugar cane '# Aibika O Each symbol represents the dominant crop in the grid section t 75m Fig. 3 Schematic diagram of a one year old swidden showing log grid and cropping arrangement (located on map 7). 52 After the swidden is cleared and burned and the grid laid, a group of men dig the yam holes using sharpened sticks. A few days later the women relatives go out as a group to plant the yam seeds. The communal garden work of clearing and digging by men and planting by women are social events with much talking and laughter and chewing betelnut. At the end of the day the individual owner of the garden is obligated to cook a meal of yams for the helpers as exchange for their labour. Men and women keep seperate gardens, but may work in each others. A male member of a woman's susu or her husband will clear her garden, build the fence around it, build her yam house, and dig the yam holes. In turn, she will plant his yams and help with the weeding, and harvesting. At harvest, the wife does the selecting of seed yams and they are stored in seperate yam houses. Food from both gardens is the domain of the wife. She alone takes yams from the yam houses for cooking. The wife has responsibility as well as power in the matter of food (Fortune 1932). When a child is born, yam seeds are set aside, and will be passed mother to daughter, father to sister's son. They are looked after by the mother until the children marry. Children are taught to garden, but responsibility for the seed yams is the mother's. When she marries, the daughter takes over for the mother and is watched closely by her in-laws to determine her competence. Nade people plant two principle types of yams; small-thorny siakutu (Dioscorea  escuelenta) and large-smooth kwatea (Dioscorea alata) (Purseglove 1972). A third and minor species of yellow yam is called bebai or noma. Kwatea grow largest and are considered the most flavourful, but grow singly and do not store well. Consequently they are planted first, harvested first and eaten first. Siakutu develop multiple tubers and keep longer. They are said to pull people through to the next harvest. Women plant the kwatea in October; they are the first crop to be planted in the newly cleared garden. Large yams are sliced and then planted, small ones are planted whole. If there are enough seed yams left from the previous years harvest they will be planted alone, 53 with no other type of crop. Women plant siakutu a few weeks later, sometimes in a separately cleared garden. The second garden may or may not be adjacent to the first depending on the soil type. Nade gardeners believe that siakutu need the sandy, tropofluvent soils of the lowlands, while kwatea need the clay loam, dystropepts and eutropepts on the slopes. They also believe that siakutu are less susceptible to competition than kwatea. Siakutu may also be planted again in the previous years garden along with other crops. They are said to be less susceptible to competition than smooth yams. The first crop in a new garden is almost pure yams planted only by women. Men and women will interplant a few banana (Musa spp.) and papaya (Carica papaya) as vegetative cuttings taken from the recently fallowed garden. After the yam harvest, in about June, men and women start planting a multi-crop phase (Fig.3). They may plant a few of the best yam seeds just harvested, along with other starchy crops such as sweet potatoes (Ipomaea batata), taro (Colocasia escuelenta). various types of banana (Musa spp.), and cassava (Manihot escuelenta) (Purseglove 1972; Williams 1979). Both men and women plant papaya (Carica papaya), pineapple (Ananas comosus). and sugar cane (Saccharum officianum) (French 1986) in the swiddens for snack foods to be eaten during a working day. These various fruits and tubers will mature sequentially for the next year. In this later stage of the swidden garden when more tall crops such as bananas, sugar cane, papaya, and cassava grow, then less weeding is required. The crops of this gardening phase will be harvested from October to June when the new yams will again be ready. Thus households always have two to three gardens at different phases of production; one developing garden (less than 1 year old), one garden at the peak of its productivity (1-2 years old), and one old garden (2-3 years old). The old garden is already in the fallow stage but gardeners may harvest the occassional tuber from it, or take out cuttings to transplant to a new garden. How Nade gardeners distribute the crops of the multi-crop phase depends on the slope and soil type within the clearing. They plant taro and pineapple lower on the slope 54 160m elevation N Forest boundary Garden division Year cleared Fencing Trail Yam house Mango tree Coconut tree Betelnut tree Piesi landmarker Fig. 4 Schematic diagram of a tala swidden cluster showing garden divisions (located on map 7). 55 where there is more moisture. Taro is also planted in rocky areas. Gardeners do recognize the relationships between plants and planting arrangement is adjusted accordingly. Cassava is recognized as a strong competitor and is mostly planted at the perimeter of the gardens. Sweet potatoes provide low ground cover making good weed deterents. Bananas and papayas are interspersed randomly and as a tall crop may act as a weed deterent. Gardeners know that taro is most susceptible to insects on the flats and grows better on the slopes. Potatoes and cassava, on the other hand are believed to grow better on the flats. Crops are generally laid out on small blocks as defined by the grid patterns of logs (Fig.3). The distribution of yam gardens depends on gardeners choices based on perceptions and interpretations of environmental conditions, labour saving strategies, and on land tenure. Site selection criteria will be discussed in detail in Chapters 7 and 8. Here it is sufficient to describe the general spatial distribution (Map 4). Each susu has a limited area of land ('watobwa) on which to make gardens, and perform other subsistence activities such as gathering construction materials, firewood, and bush foods. Susu territories usually have natural boundaries such as rivers or ridgetops. Individuals will rotate their yam gardens and fallows within that territory. Gardeners will sometimes clear gardens on their own or several members of a susu will clear one swidden and the individuals will mark their gardens within the swidden. The result is a cluster of new gardens and young fallows called tola (Fig.4). Tala range in size from 0.6 ha to 3.0 ha. II. Fallow sites The people of Nade distinguish between three stages of fallow. Yakwala refers to a recently abandoned garden with pioneer species and some remnant crops. Yakwayakwala describes the herbaceous and shrubby characteristic of fallows up to about 6 years. Waudala describes the woody forest-like landscape of older fallows, that might also be called secondary forest. Fallow gardens are dispersed throughout the flats and up the slope to 300m elevation. 56 When gardeners abandon their 18 month to two year old swidden to fallow a different vegetation structure will develop depending on the geomorphology, topography, and soil type of the site. Rate of regrowth, the type of species, and the proportion of species types will vary site to site. This variation in the environmental conditions for regeneration of fallow vegetation, as well as the variation in age of the fallows, creates a complex mosaic of successional phases over the landscape. To differentiate the various phases in terms of temporal and environmental differences would require a more detailed quantitative vegetation analysis. This initial landscape inventory is intended only to point out patterns. Within the diversity of the fallow mosaic there is commonalities. A general characteristic of early successional forest species common to the fallow is that they are pioneer r-strategists, having a high rate of reproduction and the ability to quickly colonize a variety of habitats. The majority of species coppice, or have efficient seed dispersal and long seed dormancy. The seeds can be present in the soil for considerable time and germinate in bright light or high temperatures. These plants grow rapidly and tend to dominate large gaps of over about 500m^ (Boerboom and Weirsum 1983). Indeed, many fallow species observed at Nade are common throughout Papua New Guinea, for example Alpinia spp.. Ficus septica. Ficus copiosa, Premna integrifolia. Gnetum gnemon. Kleinhovia hospita, and Commersonia  bartramia. The most common genus is Ficus. Pterocarpus indicus is abundant in all stages of fallow, and is of particular interest both as a valuable timber species and as a nitrogen-fixing. Other leguminous trees, abundant in fallows at Nade, are locally called kwa'usa, kepo'u, and maluwana. But the pattern of natural succession is complicated by cultural modifications. One hundred different plants found in the fallow mosaic have known uses, from construction materials, to foods, to medicines. Women gather bush foods from the leaves and fruits of trees. Men cut whole live trees or tree limbs for construction materials or take dead snags for firewood. Some fallow areas closest to villages are noticeably less woody for this reason. In a process of unmodified natural succession, the density of woody stems, diameter of stems, 57 heights, and canopy cover could be related to the age of fallow and site factors. In culturally modified landscapes these are less reliable indicators of recovery and stand development. III. Hamlet agroforests Surrounding the hamlets, to an approximate radius of 150m, is a cultural successional phase that could be classified hamlet agroforest. Some of these agroforests would have been swiddens at one time, though some areas may have been too rocky for yam gardening. The agroforests appear as modified fallow in that the shrub layer consists of some early successional species common to the fallow mosaic. Hamlet residents may interplant betelnut, banana, papaya, and pineapple. The tree layer is distinctive in that it consists of a higher proportion of food and multipurpose species than an unmanaged fallow garden. Both the boundary between the agroforest and the less managed fallow garden or secondary forest, and the boundary between the hamlet agroforests and the highly cultivated vegetation zones around the housesites are not easily distinguished. The size and the pattern of hamlet agroforests varies. Some have distinctive zones of palms, or fruit trees while others are more randomly mixed. The size of agroforest area seems to depend on hamlet size and location. People manage the agroforest for a diverse set of human needs. At the same time they are maintaining a plant community with a complex ecological structure. The wild and weedy herb layer provides medicines and some leafy green foods. The tree component plays a more productive role in providing foods and other materials. Some common food tree species in this zone include breadfruit (Arctocarpus spp.), mango (Mangifera sp.), guava (Psidium guajava"). ocary nuts (Terminalia koembachii'). Malayan apples (Szygium moUucanas), Ficus copiosa, Gnetum gnemon, coconut and betelnut (Areca catechu-). Various Pandanus species are also common in the agroforest, kept for obtaining fibre materials. Baubau Bambusa sp. are propogated for construction purposes. The agroforests have complicated histories due to the range of activities of hamlet residents. People select plants for their utility as foods, medicines, and sources of raw 58 materials. They may plant or transplant useful tree species, they may retain fruit trees from the swidden-fallow that occupied the site previously, or they may encourage naturally regenerating seedlings of useful species. Seeds may also germinate from household refuse. People locate useful plants, by planting or retention, where they can give them special attention or protection. But, not all plants are consciously managed. It was difficult to determine the history of some plants. People do not cultivate any staples in this zone, probably .because of the difficulty of protecting them from wild and domestic pigs. The hamlet agroforests are mosaics of plants of different ages. Significantly, there is no one perennial species that dominates the agroforest. The multi-story structure of the hamlet agroforests is similar to some natural forest ecosystems. The upper strata, up to 35m height, includes Arctocarpus spp.. Mangifera spp., Pterocarpus indicus. or Alstonia scholaris. A lower strata between 15 and 20m includes various Ficus spp., and below that the smaller fruit trees such as Szygium, and Guava. The ground layer is covered with Alpinia spp. and other early successional species. Emerging replacement trees are present in each layer. Hamlet agroforests are almost continous following the chain of hamlets along the coast. Tenure boundaries between each hamlet agroforest are marked by creeks or planted Piesi (Cordyline terminalis). IV. Coconut and betelnut groves The people of Nade integrate betelnut (Areca catechu) into the hamlet agroforests, or they propagate betelnut in small groves near the hamlets, or near heavily gardened areas on the footslopes. They believe betelnut grows best on the cool, moist, but well-drained soils of the slopes. However, betelnut is highly valued and is more easily guarded and harvested near the hamlets on the coast. Betelnut is currently the main cash crop of Nade. It is sold mostly in the Trobriands in the month of November. Perhaps because of its cash value and its social importance^ betelnut trees are jealously guarded and are often the focus of tenure disputes. 59 Nade people either plant coconut trees in groves or allow them to sprout from discarded coconuts near the hamlet or garden. Coconuts are multi-purpose, used extensively for food, drink, cooking oil, and fibre material. Many people planted coconut as a cash crop on their land in the 1970's to produce copra. Some obtained hybrid varieties from the Department of Primary Industry in Esa'Ala. But the price of copra has fallen, and transporting the copra to a market has always been a problem. Now coconut trees are planted for subsistence only. V. Mixed Sago groves The people of Nade manage mixed sago groves on the flat poorly drained coastal areas of Eastern Nade, and in a wet site in western Nade. Suitable sago habitat is limited in Nade and may account for the odd sago tree planted up to 210 metres elevation in wet creek sides. Most of the sago in Nade is vegetatively propagated to form a lower tier under a dominant canopy of trees. The large trees commonly associated with the sago are all tolerant of wet conditions (Paijmans 1979) and include Alstonia scholaris, Planchonia papuana, Terminalia spp.. and other not yet identified trees. These trees either seed spontaneously and are selectively left, or they are occassionally transplanted. Other palms such as kakolu (used for flooring) and dibila (fibers used for tying houses), and Pandanus spp. occur with sago in. the understory. The ground layer varies with the degree of swampiness even within the relatively small area of this successional phase. Most of the area has a watertable below ground level most of the year, and there the ground layer consists of ferns, herbs of the Zingiberaceae family, and shrub pandanus. Sago is important in the economy of Nade as a construction material, used extensively for roofing, and walls. Sago also provides staple food security. Abundant food starch can be harvested^ in between yam crops or when there is a poor yam crop. Harvesting of the sago trees for starch usually occurs in the months of January - March 60 before the yam harvest. Sago reaches harvestable maturity at a minimum of eight years (Flach and Schuiling 1989). Because of the multiple and important uses of sago in the subsistence economy, and the limited habitat, individuals maintain exclusive rights to the sago trees that they plant. Tenure of sago trees will be discussed further in Chapter 8. VI.Modified small-crowned hill forest The small-crowned hill forest of Nade occurs between 400 and 700 metres elevation. The history of human disturbance in this zone is unclear. Nade elders have no detailed memory of who gardened here and when. But abandoned village sites exist up to 300m and at about 400m, suggesting that at least some of this zone was gardened on a small-scale more than 60 years ago. At a few sites, I observed Pterocarpus indicus, 35m in height, growing in unnaturally straight lines at right angles - probably fences at one time. Construction materials have been removed selectively on a small scale until the present. This phase can be described as a forest that has the structural characteristics of untouched forest, where the human impacts have probably not differed in character and scale from other natural agents of disturbance. Trees are large (up to 150 cm DBH) and many are buttressed. Lianas are abundant. The herb layer is sparse and dominated by ferns. The shrub layer is dominated by small palms. The canopy is 80% closed and heights reach about 40 metres. Dominant species include Pterocarpus indicus, Terminalia spp., Dillenia  castanaefolia, Ficus spp., Albizzia falcatoria. Only some Nade men venture occasionally into this zone. The primary incentive to spend the energy to climb the hill to this area is to hunt wild pig or to obtain special construction materials for canoe building or drum carving or for houseposts. The cuscus, Phalanger orientalis intercastellanus (Laurie and Hill 1954) nests in cavities in large trees in this phase. Hunters capture cuscus by felling the trees (up to 60cm DBH) in which they are nested. 61 The few successful hunters of Nade (all men) seem to enjoy going to this zone because it is cooler and quieter than the gardened zone. The endemic Goldie's bird of paradise Paradisea  decora inhabits this forest type. VII. Modified lowland forest Structurally intact lowland forest occurs in patches of the lowlands and footslopes of eastern Nade up to an elevation of about 200 m. Again this phase shares characteristics of undisturbed forest It is characterized by trees to heights of 40m and diameters of 150cm, with an 80% canopy closure. A heavy vine layer and the dominance of small palms in the shrub layer may be indicators of some intermittent, small-scale disturbance. Various medicinal plants, and edible ferns are gathered in this zone, as well as poisons for fishing and hunting. The larger trees are suitable for construction. Resin is obtained from some tree species for glue for making drums and repairing canoes. The cuscus is also hunted by felling trees in this phase. VIE. Primary montane forest. Above 700m elevation is a small-stemmed cloud forest. Stem diameters reach a maximum of 25cm and heights reach to 15m. Mosses and epiphytes are abundant. Ferns and small palms form the herb and shrub layer. Tree ferns are common. A few Nade men visit this forest zone to hunt wild pig. Partly because of the distance from settlements, local people, except for hunters, have little reason to venture to the montane forest. A few Nade men visit this forest zone to hunt wild pig. In the past, when settlements were common at 300-500 metres elevation, the montane forest may have been a source of medicinal plants, gathered foods, and construction materials. The useful montane plants are no longer well known. FX. Nipa palm swamp 62 A small 1 ha Nipa palm (Nypa fruticans) swamp occurs to the far west of Nade. Nade people tell a legend that the palms were brought by a woman of the pigeon clan from Lake Lavu. Her son wanted to be able to fish so she brought Nipa palm down to the coast. It is said the area used to be dry and dominated by Pterocarpus indicus: now it is swamp because of the Nipa palm. Nipa palm fronds are sometimes used as a substitute for sago thatching. X. Mangrove swamps Mangrove swamps, dominated by Rhizophora spp. and Brugueira spp., occur along the coast, and up to 150m inland at Eastern Nade. The local people do not significantly modify the vegetation of mangrove forests, but they do collect clams from the mud between the knee roots of the trees. The stems of some mangrove species are sometimes used for houseposts. Saltwater crocodiles (Crocodylus porosus) are reported to live in the brackish creeks of the mangrove swamps and Nade people tell stories and myths involving the crocodile. XI. Grasslands On the drier vitrandept soils (of volcanic origin) of Eastern Nade there is one hectare of grassland. This area was probably gardened long ago (elders don't remember), and forest succession was deflected. Interestingly, there is extensive grassland on the volcanic fan fifteen kilometers to the east of Nade, and on Normanby Island across the channel from Nade. In those areas all the gardens are cleared from grasslands rather than forest-fallow. Like many areas of Papua New Guinea, the spread of kunai grass (Imperata cylindrica) is probably associated with soil degradation from repeated clearing and burning for gardens and for hunting. The presence of this grassland might indicate a susceptibility of the vitrandept soils to over-cropping. 63 XII. Beach Some plant species occur only on the sandy beach fringe. This zone is subject to high tides and some wave action as the beach erodes. An informant (Aiseya, personal communication) claims some people have planted trees for the purpose of preventing erosion. There is some gathering of medicinal herbs and shrubs in this zone. Several species of Pandanus are found here and used for fibre. There is some overlap of the beach zone and housesites as many hamlets are located 10-20m from the beach. Near the beach hamlets are nut trees such as Inocarpus edulis and others not botanically identified. XILI. Hamlet-sites The vegetation community at the hamlet-sites borders the hamlet agroforest. Every plant near the houses (to a radius of about 10m) has been selectively propogated. In between the plants the area is carefully weeded and swept clear to bare soil. Cuttings from various forest or successional phases are planted for their ornamental or medicinal properties. Such as Barringtonia papuana. Alpinia spp., Crinum asiaticum, Cyathula prostrata, and Eudio  hortensis. Specialty foods are planted close to the house for protection such as Citrus spp., betelnut, papaya, and Capsicum sp. Some ornamentals such as marigolds, and zinnias are recent introductions from missionaries. XIV. Cocoa and coffee Some local people planted a coastal variety of coffee as a cash crop in the 1970's, but they have no local market for it and transport to Alotau is difficult to organize and expensive. A few remnant plants are left. One individual planted a one hectare plantation of cocoa with a shade crop of Gliricidia in 1989, but it failed to establish. A few remnant plants are left. Over the last twenty years Nade residents have tried intermittently and unsuccessfully to establish 64 plantations of cash crops such as cocoa and coffee. The transportation and marketing of the crops are the inhibiting factors. Organization within the community, land disputes, and jealousies have also posed obstacles. A small cocoa drying and processing plant operates at Budoya, but it buys mostly from small cocoa plantations up the Salamo river, and it too has difficulties with marketing. XV. Riparian Strips of riparian vegetation are left relatively untouched. Creek sides are not cleared for gardening due to the steep slopes, and/or the rocky outcrops and rocky alluvial material. A few ornamental plants are found in this habitat - Hoya sp.; and some magic and medicinal plants - edonene (Neonauclea chalmensi) and deba'o (Crataeva religiosa). 7.2 The influence of environmental parameters in the Nade landscape. The distribution of garden and forest types in the Nade landscape can be related directly and indirectly to environmental factors such as geomOrphology, soils, and topography. But as a culturally modified landscape, the distribution of successional phases is as much a result of human management decisions, currently and historically, as environmental determinants. Environmental parameters influence the cultural landscape of Nade by 1) limiting management options and 2) influencing perceptions of the environmental conditions and thereby directing land-use decisions. Thirteen of the fifteen phases identified (Map 4 and 5 and Appendix H) are culturally modified in some way. The kind of modification partly depends on Nade people's choices based on their perception of environmental parameters. The specific result of the modifications depends on the environmental parameters of the site. The successional phases have been defined physiognomically. Each phase constitutes a complex of plants which is to a greater or lesser degree modified from a natural successional phase. In some phases, the floristic composition and vegetation structure correspond predictably to environmental parameters such as elevation, topography, and soil type - predictably, in the sense that the phase appears to be following a process close to natural succession (a process determined only by environmental variables and natural disturbances). The vegetation of other phases is altered to a greater extent by cultural activities, such as forest clearing, weeding, and planting - disturbances which will modify temporarily, or more permanently, the process of natural succession. Two of the fifteen phases are not modified due to environmental limitations. Nade gardeners never clear the steep riparian ecosystems (though one case was observed where swidden clearing was too close to the edge of the creekbank causing a small landslide into the creek); nor do they clear the small-stemmed montane forest. Outcrops of rock prevent cultivation of some sites near creeks. People have never cleared gardens on some of the extremely rocky alluvial areas on the coastal plain. This explains the existence of small remnant patches of mature lowland forest. But because of the proximity of these areas to hamlets, people continually harvest various products from them, modifying the forest structure somewhat. Some plant communities, such as sago groves, nipa palm, and mangrove swamps, are adapted to very specific substrates, and microenvironments. They in turn, create specific habitats. Sago groves, for instance, exist only on the poorly drained coastal plain. Under population pressure, these environmentally selective phases would tend to be the most susceptible to over-use, mismanagement, and disputes. The pattern of the forest-fallow mosaic is largely determined by the swidden site selection, which in turn is partly determined by Nade gardeners perception of environmental parameters. Nade gardeners have little choice but to make their yam gardens on some very steep slopes (up to 60%). But they recognize some advantages of slopes in terms of drainage and the ease of clearing the bush. Gardeners also consider elevation when selecting a garden site. High gardens (between 300 and 450m) involve more labour in terms of time spent 66 travelling to them from the hamlet and in clearing the mature forest; on the other hand, gardeners expect a higher crop yield from the rarely cultivated soils at higher elevations. Some gardeners claim that the soil is better for yam gardening on the slopes; it is cooler and wetter for growing and softer and easier to work. Near the bigger creeks the soil is said to be even cooler and moister. People all believe the soil is better where there has been big trees. When asked why, the answers varied: 1) soil is cooler and moister 2) soil is fertilized by the falling leaves 3) if trees grow big then the soil must be good In terms of soils most local people categorize four areas. The flats, the slopes, west of Bwaubwau Creek, and east of Bwaubwau. Bwaubwau Creek is indeed where there is a transition from the dystropept and eutropept soils over metamorphic rock, to the vitrandept soils of volcanic origin (Map 2). Yam gardens and the subsequent fallows are created on a wide range of geomorphological, soil, and topographical conditions. The rate of regrowth and the fallow components would depend on the structure of the plant community prior to clearing, the surrounding vegetation, topography and soil (Ellen 1978). The fallow species differ noticeably on the vitrandept soils of eastern Nade, and the dystropept/eutropept soils of the slopes of western Nade. Strictly on the basis of environmental limitations there are more management options for the fallow areas than for sago groves, or mangrove. Fallow can be cleared for yam gardens, gradually converted to hamlet agroforest, or left to become mature forest. 7.3 The influence of management techniques on the Nade landscape The cultural successional phases are essentially a spectrum of cultivated, semi-managed to modified-wild management zones. This spectrum, representing an integrated management strategy, is typical of many traditional land-use systems in tropical rainforest. The managment of individual and groups of plants creates patterns that overlay the mosaic of spontaneous regrowth caused by the variation in microenvironments. In terms of individual plants, any phase at any one time is comprised of a certain proportion of planted, or transplanted plants and untended weedy or wild plants that have been slated for removal or ignored. The successional phase can be placed on the management spectrum by comparing the proportion of managed, or selected plants and the non-managed components. Housesites have the highest proportion of manipulated plants. Ornamental plants and fruit bearing plants, mostly introduced species, are planted around houses. The ground between the cultivated plants is weeded and swept clear of any herbs or grasses. The soil is generally compacted, altering the growing site significantly. Human disturbance to the biophysical system (vegetation, fauna and soil) is more intense than any other phase, and can span an extended period of time. Yam gardens have a high number of manipulated plants; those that are planted, transplanted or tended (ie.pruned or weeded to alter growing conditions). They are managed for short-term productivity. They require the most energy input in terms of labour. Ultimately, the long-term role of these sites in the landscape will be determined by the interval between swidden clearing and clearing size, which in turn influences soil dynamics, microenvironment and the plants that are able to recolonize. The en masse vegetation clearing for gardens, and the weeding and slashing back during the 1-2 years of yam gardening affect the fallow cycle. This kind of vegetation disturbance will have different effects on different species. Some species will die, some will coppice, some will quickly invade the cleared site. Frequent slashing can lead to the local extinction of some species that are unable to adapt to such manipulation (Alcorn 1981). Agricultural techniques in the yam garden, such as selectively weeding tree seedlings while 68 pulling out herbaceous weeds, affects the long-term development of the landscape by altering the fallow dynamics. In general, the forest-fallow mosaic at Nade appears to be similar to the highly productive early successional growth phase of a natural forest. Fallows are more or less modified depending on the proximity to settlements. There are no tended or cultivated plants in this phase. But there is the greatest diversity of useful plants. Individual plant management includes: harvesting of fruits, young leaves, or roots, or complete removal of trees for construction material. Manipulation of individual plants during the fallow is probably insufficient to significantly alter the process of natural succession. Harvesting is often an incidental action in that people will pick some fruits or greens while returning from the yam garden, rather than making a special trip for the purpose of harvesting. Energy input is minimal. The intensity and duration of human disturbance is similar to natural disturbance. The most important management decisions affecting the structure and function of the fallow mosaic are fourfold; (1) how often the fallow is cleared for gardening; (2) what type of vegetation community is cleared; (3) the length of the crop rotation; and (4) how the garden is weeded. It is the frequency and intensity of disturbance which determine the successional process by determining what species are able to colonize. Informants stated that 7 years is a typical length of fallow. However, many sites are left up to 40 years. Generally, the nearer the garden to the coast and to hamlets the shorter the fallow. Many Melanesian swidden cultivators actively encourage regrowth of trees in the fallow (Clarke 1976; 1991). Management techniques used to favour forest regrowth include: (l)few and selective weedings; (2)short cropping; (3)planting trees and (4)prevention of fire. Infrequent weeding during the cropping phase, and short crop rotations are related. A second cropping in immediate succession usually necessitates increased weeding, which lessens the store of tree seeds and seedlings and reduces the possibility of primary forest species regenerating in the initial regrowth (Gomez-pompa et al. 1972; Unruh 1990). 69 The people of Nade do see ecological benefits in the forest-fallow phase, noting specifically the importance of trees. They have an indirect awareness of the value of the fallow stage for soil health. They claim that the productiveness of their garden depends on the type of vegetation from which they cut it. They understand that the longer the fallow is left, as measured by the woodiness of the fallow phase, the better yields they will get from their gardens. However, they do not seem to recognize any particular species as important indicators of a healthy fallow. I was unable to determine to what extent forest fallow regrowth is actively encouraged by the gardeners of Nade. Although they are aware of the benefits of fallows their awareness is not necessarily translated into active intervention. The Nade gardening techniques that would affect the structure and composition of the fallow regrowth are the selection of species for living fences and yam posts and some selective weeding of the mixed crop gardens to favour tree species over herbaceous species. Later in the fallow cycle, Nade people sometimes create gaps in the regrowth by removing trees for construction materials. The resulting micro-environment heterogeneity in the modified fallows would create variable conditions for the germination, seedling establishment, and growth of plants from distinct successional states (Brokaw 1980; Gomez-Pompa 1972; Whitmore 1975; Unruh 1990). Even swidden cultivators who appreciate the benefits of a long forest fallow may not want to allow a plant community resembling "primary" forest to develop. At Nade, secondary forest and edges of primary forest have more useful species than the "primary" forest itself. Secondary forest is easier to clear and offers more useful species for fuel, building materials, food and medicines. The increase in crop yield from fallowing soils starts to flatten out after 5-10 years of regrowth (Clarke 1976). Thus there is no advantage for a subsistence gardener to allow the succession to approach climax. The problem of sustaining swidden cultivation in tropical rainforests is one of allowing succession to reach reasonably well-developed secondary forest, while preventing degradation to scrub or grassland. At 70 Nade, the fallow appears to be sufficiently long and the cropping cycle short for crop yields to be maintained. However, the spread of anthropogenic grasslands in some areas surrounding Nade is extensive and the direct cause has not been investigated. Environmental conditions (climate and soil type) and land use pressures (resulting in shortened fallows without fallow enrichment, and repeated burning to enhance opportunities for hunting) are probable causes. The hamlet agroforests of Nade have a lower proportion of manipulated plant species than yam gardens and more than the fallows. The vegetation structure is similar to, and some plant components identical to, a mature fallow phase. Many plants are manipulated intentionally by transplanting seedlings or cuttings, or incidentally, by pruning for firewood or harvesting greens. Ferns and firewood may be gathered on the way back from the yam garden; a seedling of a fruit tree may be noticed and left; and firewood may be cleared from around it to give it more light. Energy input is much less than for yam gardens and is often incidental. The plant manipulation in the agroforest has a more long-term effect on the process of succession than the human disturbance in the fallows. Like the fallow mosaic, the hamlet agroforests are apparently derived from the surrounding tropical forests. They would be gradually transformed from a fallow or secondary forest structure by the selection of useful species, elimination of useless species, and introduction of new species. The forest-like structure could be attributed to a lack of any regular planting pattern and the incorporation, or rather allowing, of successional forest tree species. Whatever the route taken from the original forest to more artificial agroforests, it has involved a modification of the ecological and biological features of the wild useful species as well as an adaptation of the characteristics and mechanisms underlying the functioning of forest ecosystems in order to produce an environment according to people's needs (Michon et al 1983:122). In sago groves there are fewer tended plant species, at any one time, than a yam garden. The floristic composition of the phase is directly influenced by the maintenance of sago trees for the duration of their 8-15 year rotation. Disturbance is less intense than yam 71 I gardening, but more continuous. Disturbance is on the scale of small natural gaps. The soils are never completely exposed. However continuous selection of species may, like the hamlet agroforests, result in a gradual transformation of the plant community. Some plants are selectively favoured, while some original plant components, with inefficient dispersal mechanisms may not be able to recolonize. 7.5 The agroforestry landscape at Nade Agroforestry is a collective name for land use systems and technologies in which woody perennials (trees, shrubs, palms, bamboos) are deliberately combined on the same management unit with herbaceous crops and/or animals either in some form of spatial arrangement or temporal sequence. In agroforestry systems there are both ecological and economic interactions among the different components (Lundgren 1982:3). According to this definition, the landscape mosaic described in this study can be called an agroforestry system. Although no one phase resembles the neat, modern conception of agroforestry with trees and annuals intercropped in rows, the land management system as a whole consists of tree crops and annual food crops combined either spatially or temporally. Almost every successional phase identified has something to offer the subsistence economy. In combination, they provide a continuous harvest of products and are ecologically integrated. The economic relationship between the phases is obvious. Subsistence products ranging from tuber staples to fruits, nuts, greens, firewood, construction, materials, medicines, body decorations, perfumes, tools, and cordage are available, at various times of the year, in the different phases. Some products are available year round while others are seasonal. Products in any of the subsistence needs categories - food, construction materials, medicines - are available from one or more successional phases at any one time. Maintaining a diversity of products, in a wide range of landscape types, lowers risk and insures all subsistence needs are met. Ecological relationships between the modified successional phases of the Nade landscape are complex, involving both spatial and temporal relationships. Two or more 72 phases may host common plant species, simultaneously at their spatial transition, or sequentially as one phase transforms into another. Some phases of the agroforestry landscape are, in fact, a more or less managed succession of another phase. The manipulation of plants in one phase will have an affect on another by creating or destroying the conditions for growth for certain subsequent plants. In other words, there is a temporal relationship between the successional phases, most evident in yam gardens, fallow mosaic, hamlet agroforest, and mature forest. Yam gardens are left to phase into either a fallow garden followed by secondary forest, or if close to the hamlet may be transformed into a hamlet agroforest. The transitions between phases will vary in time. A fallow zone may be left for five years and then be cleared for a yam garden, or a fallow may be left for thirty years to become a secondary forest zone. The spatial relationship is observed at the boundaries, for instance, between hamlet agroforest, and sago grove, or between fallow gardens and hamlet agroforest, or between fallow gardens and mature forest. It is the tree species within each phase, that because of the length of their growing cycle, are often the common component shared by two or more phases. In the polyphase agroforestry landscape it is trees (that) provide the richly productive and protective matrix that (is) also a part of each people's cultural heritage, not only materially but also spiritually (Clarke 1991). Yam gardens and hamlet agroforests represent an agroforestry system in themselves, with a spatial relationship of annual and perennial crops. The yam gardens are managed with a multi-crop phase in which trees represent a small, but functionally important, component. Banana and papaya trees efficiently utilize vertical space while contributing food production, and apparently without competing with the tuber crops. Pterocarpus indicus. as a living yam post and fence, may be important as a nitrogen fixing species and for providing shade, while it is also an important tree for construction materials. Living fences are also made from Pterocarpus indicus and other coppicers. Note that it is not known how efficient Pterocarpus indicus is as a nitrogen fixer, but trials are being done at the Forest Research Institute in Lae (Neville Howcroft, personal communication). Hamlet agroforests represent an agroforestry phase of greater structural and functional complexity than the yarn gardens. Hamlet agroforests are characterized by both cultivated plants and forest species and could be described as a highly modified succession of the yam gardens. Hamlet agroforests are a more permanent type of land use (as long as the hamlet remains) which provide a range of products such as fruits, nuts, firewood, timber, ornamental, and medicinal plants - convenient to residents when desired. The structure of the agroforest is layered like a "natural" forest with vines, lianas, herbs, shrubs, and trees. The structure is a continuously changing dynamic. The permanency of production results from a constant pool of emergent useful plants. These are renewed by the interplay of natural processes of plant reproduction with introduced human practices, needs, and customs. The forest-fallow mosaic also performs important economic and ecological functions in the agroforestry landscape. Of all the phases, this phase supplies the greatest diversity of minor products. Moreover, as the fallow undergoes natural succession toward mature forest, it also provides hydrologic benefits, microclimatic modification, soil erosion control, soil fertility enhancement, and wildlife habitat. It represents an intermediate environment between clearings and mature forest. In the Nade agroforestry landscape, the mature forest phases provide some food and construction products. In addition, this "deep" forest or uwama engenders respect by people of Nade who believe it hosts forest spirits both good and evil. It encompasses their sacred sites and their mythologies. Further anthropological research may find that although uwama is less frequently used and visited than the other successional phases, it is very significant in the Nadean perception of their environment. The modified lowland forest and small-crowned hill forest phases have important ecological functions in the Nade agroforestry landscape including: regulation of 74 microenvironments, habitat for rainforest animals, a source of seeds for recolonization of rainforest species in adjacent secondary successions, and a source of genetic resources. Mature rainforest is a mosaic of heterogenous plant communities created by gaps in the canopy caused by small-scale natural disturbances (Brokaw 1985). The size and frequency of these distubances will influence the composition of regeneration (Denslow 1980). Many of the plant species of the mature forest have a low rate of reproduction and propagule dispersion and little or no seed dormancy. These species are mostly shade tolerant as young plants and can germinate only in undisturbed forest or in small gaps. Many rainforest species are dependent on specific pollinators, or on certain vertebrates for seed dispersal (Whitmore 1984). These species characteristics are part of what makes mature tropical moist forests particularly vulnerable to disturbance. The high diversity and the structural characteristics of tropical rainforests create alternative paths for the movement of matter and energy, in turn enhancing the internal stability of the system. However, the long evolution, under relatively stable conditions, of high degrees of specialization and low rates of reproduction and dispersal make the system susceptible to the larger and more frequent forms of disturbance introduced by humans (Boerboom and Wiersum 1983). To sustain the biological diversity of the agroforestry landscape, a sufficient area of mature forest fragments must be maintained or allowed to recover. Early successional fallow phases are characterized by relatively simple food webs. The smaller set of niches may not promote the same levels of species diversity as a landscape mosaic that includes tracts of mature forest (Ingram 1989). The size and distribution of mature forest patches in the landscape, and the size and distribution of swidden clearings determine, to some extent, the ratio of pioneer species and the primary forest species able to recolonize in the fallow mosaic. Repeated disturbance which precludes the retention or establishment of mature or climax vegetation can lead to progressive depletion bf species (Peet et al. 1983). 75 Given adequate fallow, a swidden-fallow cycle is effectively a sustained yield forest management system. Moreover, under ideal conditions a swidden system of cultivation could maintain the overall biodiversity and ecological structure of a natural forest landscape. Under swidden cultivation land is exhausted at a faster rate than it recovers, so the area of regenerating land must be several times greater than the area being cultivated if new land is not to be used (Carneiro 1960:229). The ratio of yam-garden area to fallow area and rates of regrowth can be used as indicators of sustained yield, but more factors are involved in conserving biodiversity. The size and distribution of mature forest phases and the size and distribution of the individual swidden clearings over the landscape will affect microenvironments and the types and distribution of habitat for both flora and fauna species. In Nade, the most significant forseeable change in the ecology of the agroforestry landscape would be an increase in the percentage of land used for perennial cash crops. The land area used to grow perennial crops such as coconut, betelnut, cocoa, coffee or oil palm is taken out of the swidden-fallow cycle, and not allowed to revert to a mature forest phase. As the percentage of land area under tree groves increases beyond a point of balance (as yet undetermined) with the swidden-fallow cycle, the biodiversity over the landscape would decrease. 76 Chanter 8. T H E SOCIAL ORGANIZATION OF LAND: CUSTOMARY LAND  TENURE. AND LAND USE DECISION-MAKING. The functioning of a subsistence land use system cannot be described without some reference to social organization. Individual decisions involving land and plant use are made in the context of social and political structures, and rights and obligations of members of the group as a whole. As Malinowski observed with the Northern Massim people of the Trobriand Islands, Land tenure enters very deeply into every aspect of human life, and it is the integral expression of all the ways in which man uses his land and surrounds it with values of avarice, sentiment, mysticism, and tradition (Malinowski 1935:319). The description and analysis of land tenure in traditional societies is frought with difficulties. This is partly due to the cultural bias of investigators who use the terminology of the industrialized world's legal and economic system. A detailed mapped survey of the Nade land tenure system is beyond the scope of this study. Rather, the objective is to understand the kinds of land and plant rights, and the modes of transfer of rights in order to determine the influence of tenure on land-use decision-making and thereby its influence on the forest landscape. 8.1 The customary tenure system at Nade. 8.1.1 The land tenure system Nade ward is broken into several units of land each under the exclusive claim of the susu that originally cleared it (probably around the turn of the century when clans from the interior of the island settled at Nade). The susu maintains a claim to the land for as long as they can trace their ancestoral ties to it. Members of the susu have rights to cultivate gardens, hunt, collect forest plants, and build residences on the susu land. Heirs should be able to recount their genealogical connection to the founder or original claimant (do'a) and they can name plots, creeks and other boundary markers defining the area. 77 The normative system of kinship inheritance follows the matrilineal line. An individual's rights to land come to him/her through the mother. When a couple marry their children garden on the husband or wife's land. When the parents die the children maintain a right to garden on their mother's land, but lose the right to garden on their father's land. The father's land will be used by his sister's children; it remains the land of his susu. Land is never actually apportioned to heirs, rather they inherit the rights to use susu land. All land is ultimately susu land; but one susu member does not have the authority to acquire or dispose of land independently of the susu. Individuals make the day to day decisions regarding land use on their own, or in consultation with susu elders or other members. In a sense, the interests of the susu and the individual are identical and there can be no contradiction. Individuals have the right to use a plot of susu land for a yam garden. The food a person grows there is his/hers as a product of his/her labour and seed source. While there is yams in the garden it is a place of privacy. When the owner is not in the garden it is forbidden for anyone outside of his/her susu to enter, or even look at it closely. On abandonment of a cultivated site to fallow, the land is open to use by any other member of the susu. The strict inheritance of land rights is by matrilineal descent, but there are some alternatives for transfering land rights which add some flexibility to the tenure system. Land rights may be given as payment for a service. A custom, shared with the people of Morima to the East, is that if a family helps to make a funeral feast for another they may be compensated with land. Land rights may also be given away to someone of another susu if the recipients do a significant service for the owner susu, such as caring for a child or an old person, or helping the owners to protect the territory from attackers. These gift-rights are said to be permanent, but in later generations they are commonly disputed. Land may also be passed between susus sharing the same bird ancestress. Essentially, this is a gift for kin in need. According to oral history, when the Salakahedi 7 8 people fled from tribal fighting in the interior and came down to the coast, a Nade susu with the same bird "adopted" the immigrants and gave them land to use. Rights to land may also be given for a price - traditionally yams or pigs. Such an arrangement is seen as temporary, something like a lease. To maintain their rights the susu or individual must pay compensation for the land every new gardening season with yams or pigs. Here the agreement can be with an individual if it only concerns temporary rights to the land. It may be that the purpose of continuous payment of food is akin to renewing a lease. It is a way of remembering the agreement as it is passed through generations. Such payment for land use is not a common occurence. It is probably done when a susu does not have enough land or to accomodate an outsider. In other words, there is not a concept of wealth of land. Land is not rented or traded for the sake of acquiring and accumulating it. Land transfers are part of the system of general reciprocity, and serve to maintain social relations between susus through a web of rights and obligations. The practice of "adoption" by incorporating immigrant families sharing the same bird, or individuals by way of the namesake system^, increases the fluidity of the inheritance system. If susus are diminishing in size or lose female members to pass land through they may adopt members. A man who has only brothers may adopt his own children into his susu by paying compensation usually with yams or pigs or service to his own susu. His daughters or sons gain use rights to the land. On his death compensation should continue to be paid to maintain the usufruct rights. Modes of land transfer other than through strict matrilineal inheritance add flexibility to the system allowing for more equitable land distribution between susus. It makes it possible to reapportion land in each generation according to the expansion and contraction of land-holding groups. Land transactions represent a continuing relationship between lineages rather than an instant procedure. However, lands transfered via these alternative modes are the most common areas of dispute. People forget why the land was given a generation ago and under what terms. 79 8.1.2 The plant tenure system Some plant species are maintained under a system of rights. The plant species involved and the type of rights differ depending on the context. The annuals planted in a person's garden are the exclusive property of the planter, through to harvest and consumption. Yams are the only crop for which the seeds themselves are also considered personal property and passed through generations. Planted or transplanted trees, valuable for fruit or timber, belong to the planter who has exclusive access to them. Some planted food trees to which property rights apply include: sago, breadfruit, mango, citrus, betelnut, coconut, ocary nut, and chestnut^. Sago and bamboo are planted to ensure a supply of construction materials and they also have rights applied. Trees with both high value and distinctive form, such as mango and coconut, are planted as landmarkers for garden land and at hamlet sites. Piesi CCordyline terminalis) is the most important land marker plant used in Nade, as it is throughout Papua New Guinea. On the death of the planter rights to planted living trees and their produce are transfered to offspring in the matrilineal line, or to the susu in general. There is a taboo in Dobu and Nade which applies to the children and sister's children of a deceased man, that they must abstain from eating the fruits of his trees that are planted near the hamlet or the graveyard***. This taboo seems to be less strict for trees planted in gardens distant from the hamlet. The taboo serves to maintain the trees and produce thereof in the mother's susu. Rights to wild plants, incidentally growing on a clan's land, seem to be less explicit, and sometimes ignored. One individual with land rights to most of the lowlands where a particular edible fern grows, is denying others the right to pick those ferns. He is generally ignored. Large non-food forest trees are sometimes felled to catch cuscus regardless of whose land it is. Wild trees to which exclusive rights apply include Gau, a locally rare tree of the Cinamomum genus, used for carving and magic, wild mango and citrus trees and some 80 species of Pandanus. At abandoned hamlet sites, both wild and propagated plants are left untouched for their sacredness, rather than the exclusivity of use. Another means of maintaining exclusive use of plants is through access to information. Many medicinal plants and plants used for magic are secret to a particular susu, or to specialists within a susu. Though the plant may be a common forest plant, its use will be kept secret. A nickname will be created for the plant to hide its identity in conversation. Rights to plants allow owners to demand compensation for damages to their plant property. People are not allowed to walk through other's gardens without asking permission. Owners are distrustful. Such a trespasser could be stealing plants or putting magic on the garden to make yams walk away, or not grow He or she may be asked to pay compensation of yams or live with the threat of retaliation with magic. If accidental damage is done to property (soil, plants, fence, pigs) the owners will demand compensation. An example of this is the story of a man from the east of Nade, who with entrepeneurial inspiration and a loan through the Agricultural Development Bank, bought some cows. He did not feed them well enough so they ventured into another susu's garden for greens. The susu complained to him with no results, so they filed a complaint with the Department of Primary Industry (DPI) in Esa'Alsa and the man was told to compensate them lOt for each plant eaten by the cows. Already in debt from buying the cows he could not pay compensation, nor could he feed the cows better. The cows kept eating plantings so the susu was given permission by the DPI to kill the cows and keep the meat. DPI was asked to mediate in this dispute because it involved an agricultural innovation facilitated by DPI. Property damage involving only yams would likely be dealt with within the village. 8.1.3 Sources of tenure disputes Land disputes often go back to an argument over who were the original settlers to first clear land in the area. Those who arrived from Goodenough and Normanby Islands two 81 to three generations ago are considered immigrants, though there is some disagreement about when the different susus arrived and which are the so-called irnmigrants. The original susus adopted irnmigrant families sharing the same bird and gave them land to garden on. The irnmigrants have usufruct rights to the land. Problems arise when the irnmigrants, after a couple of generations, start claiming more of the susus' land. Or the original owners recognize the cash value of the land and reassert their claim to it. The two most current and outstanding land disputes are over such a situation. In cases of land rights transfers that do not strictly follow matrilineal inheritance, then the agreements between susus should be maintained by oral transmission, and should involve compensation. But people don't always keep up with compensation payments and they forget the terms of agreements. Ill-defined boundaries can be the source of minor disputes. Plants may be unwittingly cut from another susu's land in the process of clearing a new garden. But the practice of susu members clearing land in clusters reduces the likelihood of these disputes arising. The densely gardened low elevation land is especially well marked with the piesi plant and fences. The private ownership of plants can be a source of conflict when trees and other perennials are cultivated. Sago and betelnut are common objects of dispute. Suitable growing sites for sago are limited. Some susus do not have any appropriate land and are given permission to grow sago on another susu's land (usually one that shares the same bird ancestress). The land is said to be borrowed and the trees belong to those who planted them. However, disputes arise after a generation or two when the original owners want to reclaim the land, and the original agreement is forgotten or challenged. For the people of Nade a dispute over land within the susu is almost a contradiction in terms. Individual and susu land are ultimately the same thing. Disputes that could possibly arise would be over the ownership of trees. Planting a grove of trees is an extended use of land that takes the land out of the flux of land use by members of a susu. When swiddens are 82 cleared, cultivated, and abandoned the labour investment is temporary. For more permanent groves the labour involved in transforming the forest and maintaining the crop is continuous. Thus there are quite frequent disputes over betelnut, and more disputes could be expected over other cash crops. 8.1.4 The resolution of tenure disputes. Some sources of land disputes have been described. It is inevitably at swidden clearing time that they fully arise. Land ownership is not mapped or documented; rights are maintained in the oral histories in the memory of the old people. Sometimes disputes are settled amicably by going back to the elders of each susu and getting the stories straight. If that is unsuccessful then those in dispute will go to the elected village councillor for mediation. The councillor will call a hearing and request the presence of both men and women elders of the disputing susus. The meeting takes place under the big mango tree at the Nade mission station. The mostly male elders speak in turn, refering to land by rivers, hills, big trees, and stones, or by local place names allocated to areas defined by natural features. They refer to stories of which susu arrived in Nade first, where from and where they landed; then which susu gave land to whom and why, how and when. Normally, the councillor and two other elected council members will hear the case. Officially, they have the authority to make a judgement. But disputes can be divisive for the whole community, and elected council members have susu loyalties, making them reluctant to force an authoritative decision. Commonly, disputes will be discussed for years (sometimes generations), directly between individuals, and at community meetings. At the disputing parties' request, the case may be presented to the land commissioner at Esa'Ala, the district capital, on nearby Normanby Island. This occurs infrequently, and even when it does a satisfactory resolution is rarely reached, bringing the dispute back to Nade again. 8.2 Vectors of change in the land tenure system. 83 According to elders there were fewer land disputes during their youth than there are now - then people abided by the custom of matrilineal inheritance or paying the necessary compensation. Although it is doubtful that land tenure was ever simple and undisputed, certainly the customary tenure system exists in different circumstances now. For to add to the vagueness of territorial boundaries maintained in oral history, and forgotten agreements, there is a breaking down of the traditional system of acquiring and passing on land. There are several reasons for this, all stemming from increasing integration in a cash economy. Traditionally it is impossible, even irrelevant, for an individual to acquire land. Land belongs to the susu. Compensation payments could be made to transfer land rights between susus via individuals of a generation. Land is not regarded as a commodity like cash. It has been involved in the traditional exchange relationships between susus similarily to the exchange of food. Land can now potentially be used to earn cash via cash-cropping production. Thus land takes on a new value. And this places new stresses on older systems of allocation and use. A potential for a shift in land allocation relates back to the matrilineal susu unit. Both Fortune (1932) and Atchison (1974) have rioted a basic tension in the susu structure. A wife is independent of her husband in that she has her own garden on her susu's land, and her own seed supply both of which will pass to her children. A man's village land, garden land, personal name, status, village palms, and fruit trees he must pass to his sister's children. But he is torn between obligations to the susu and wanting to provide for his own family-by-marraige. So, according to Fortune, a Dobuan man seeks to subvert the matrilineal line by dividing those things that are divisible, such as garden land and magic between his own children and his susu, in order not to vest all his power in the susu line. Although this tension appears to have existed for some time, a cash economy offers the opportunity the fathers seek. As migrant labourers men now have greater potential than women to buy or register their land through the government so he can plant his land with cash crops and pass it to his children. The experience in the matrilineal society at Lihir, and 84 a future possibility in Nade, is that men are using their cash incomes to purchase land for their children from the descent groups to which the fathers belong (Filer and Jackson 1989). To date, the only case of a land purchase in Nade, is land bought with cash by the current schoolteacher (a woman from the Trobriands) who is related through marraige to the selling susu. The purchased land is adjacent to the mission station, which was given to the church by the same susu. Members of the susu still maintain that the mission land is theirs while it is under use by the church for an undefined length of time. The permancy of the rights that were transfered to the schoolteacher and her children is also unclear. Three sisters of Nade, with the help of the eldest's university educated husband, are starting the process to officially survey and register their deceased father's land in their name. Their intention is then to begin a small cocoa plantation. The possibility of alienating land to non-Nade people has not yet arisen. But even the possibility of the selling and/or registering of land by individuals within Nade represents a significant change in their relationship to land and to each other. Ellen (1978) describes such a change for the Naulu people of Eastern Indonesia as the complete negation of the continuity and context of social relationships. The sale of land is not merely a material loss, it is also a denial of the value of traditional multi-stranded exchanges (Ellen 1978:102). As activities and relationships extend outward from Nade village, there are repercussions on the relationships of power and influence within Nade with regards to land. In addition to the potential for direct cash value of land, the market economy offers off-island labour opportunities which can have an effect on land allocation. Labour migration and off-island marriages result in people leaving for periods of time and not making continuous use of potential gardening land in order to maintain usufructory rights. The knowledge of customary land rights and the memory of previous agreements between susus are part of the status and power of the older people of Nade. Young people who have spent less time in the village setting in their youth due to their attendance at school, generally display less knowledge of plants and their uses, and less knowledge of landmarks, geography, and the 85 history of their own susu lands. The following experience of elders at Lihir (Filer and Jackson 1989) is common throughout Melanesia including Nade. Elders are threatened by the economic independence which young people can have from wage employment. But, to the extent that young men still need access to customary land, they are obliged to recognize the leadership of their elders. The elders' knowledge can be a tool to assert greater control over the distribution of land rights in order to compensate for the reduction in their control over other areas of traditional exchange. Cash-cropping could imply a shift toward rules of commodity property. The cash produced may or may not follow channels of traditional exchange. If market rules prevail, then the rules governing the exchange of produce of the land would probably affect attitudes to the land itself (Ellen 1978). In Nade, the potential cash-crops are tree crops. Regardless of the path of exchange of cash produce, a drive to put more land under perennial management of groves of coconut, betelnut, cocoa, coffee, or oil palm, would suggest a new relationship to land. The focus of labour investment would need to shift from the annual subsistence production of yams. Labour investment in a particular piece of land would span generations. Land use would be on a different time-scale. Land use would be less flexible, and land transfers less fluid. Ultimately, Changes in concepts of ownership and recognition of new and different ways of holding and transferring land have both social and ecological implications (Ellen 1978:107). This will be dicussed further in section 8.5. 8.3 Sacred limitations on cultivation. Gravesites have a special land status, giving them social and ecological significance. The sites are chosen by susu members, and are usually situated close to hamlet sites on susu land. Numerous gravesites are now situated near abandonned villages at higher elevations, by current hamlets on the coast, or isolated in the bush (Map 6). Their locations are local knowledge and considered sacred. The presence of immediate susu members is forbidden 86 near graves. Affines, and those who are unrelated may visit, but are forbidden to disturb the vegetation. The sites will never be used for gardening. Initially, ornamental plants may have been planted, but the sites are not maintained beyond that. Eventually the introduced plants will be growing under the regrowth of secondary forest. Abandoned hamlet sites also have a sacred status and are never cleared, perhaps because of their connection to gravesites. They are often marked by flat stones. 8.4 Land use decision-making There are a number of factors influencing garden and forest management decisions including: land tenure, plant tenure, labour-saving strategies, tradition, and perception of the environment. The single most rigid factor limiting choice of garden land is land tenure; land that is used or claimed by others. Thus the land tenure system could be seen as the framework within which land use decisions are made. How many, and what size gardens are cleared depends on the number of yam seeds left to plant, the amount of labourers available for clearing, burning, and weeding, the amount of cash income from relatives, and the land available. These factors will determine whether or not a household makes one or two gardens. Both men and women may decide on where to clear their new garden, individually or after some informal discussion with other susu members, particularly the elders. However, clearing is the men's work, and frequently the men decide where they will clear the garden for their spouse, sister, or mother. The general area is discussed with the woman relative, but the specific location is decided when he gets out there with axe and machete in hand. Within their own gardens, individuals, sometimes in consultation with their spouse, decide what plants to plant where. Swidden clearing choices are only partly dictated by making the best use of fertile land. Another consideration is to clear gardens in order to maintain usufruct rights to land. By gardening on a spouse's land but coming back frequently to garden on one's rightful susu land then one maintains one's rights and maintains one's knowledge of the susu territory. 88 In brief, an individuals choice of garden site is based on: (1) Limitations set by customary land tenure a. Land must be within susu or affine's land. b. Strategic reasons, such as making use of some susu land in order to maintain a claim. . . (2) Anticipated labour requirements a. Distance from village and elevation determines amount of time and energy required in travelling to and from garden, in maintaing the paths and in carrying loads of garden produce and tools. b. Proximity to previous years garden site allows the re-use of the yam house and partial fencing. c. Creating the minimum number of wild boundaries protects the crops from wild pigs, and weeds. d. The maturity of the fallow determines the amount of woody stems and thus the labour required to clear the garden. (3) Perception of environmental parameters. a. Soil fertility is judged by the size of fallow trees, amount of rock, slope, proximity to creeks. b. The quantity and type of yam seeds left for planting from the previous years garden dictates the clearing size and soil type needed for the new garden, and whether one or two separate gardens are needed. (4) Protection from yam thieves, and malevolent forest spirits. a. The further from the hamlets and from main trails the easier it is to avoid thieves. On the other hand there is more likelihood of trouble from malevolent forest spirits. (5) Personal preference. a. Individuals weigh the importance of each of the above differently. b. Individuals may prefer solitude; may want to avoid working near certain people; may want to work near susu members for the social interaction. Each of these criteria may take greater or lesser precedence depending on the individual and the context. In general, the garden site will be selected according to the appropriate environmental parameters for the available seed stock, within the susu block. 89 8.5 The cultural influence on the Nade agroforestry landscape. The current distribution of susu land blocks is a result of ancestral histories -movements of susus and relationships between susus over the last three to four generations. But the stories of these movements are numerous and conflicting and cannot be unravelled by this study. The distribution of swidden clearings (and thus the distribution of subsequent successional phases) over the Nade landscape, in any one year, is a combined result of choices made within each susu block. As evident in Map 4 there is a tendency to cluster gardens primarily as a result of criteria (2), (4), and (5) above. Clustering of swiddens within the forest landscape has ecological implications. In short, the size, frequency and distribution of clusters affects the local environment and the ability of different plant species to colonize, which in turn affects the movement of animal species. Susus vary in the number of members. Some susus claim more land than others, though there is not necessarily a relationship between the size of susu and the amount of land the members have access to. The result is that susus with less land may be cropping on a 5 year fallow rotation while others may be cropping on a 20 year rotation. The susus with access to the most arable land have the opportunity to make agronomic decisions about where to clear their gardens on the basis of the best soil and location. Susus with less land may be forced to garden on less appropriate soil types, or to clear younger fallows. Thus, the length of the fallow cycle depends on human population dynamics, and land tenure constraints. If the people of Nade village were to maintain land rights strictly within formal kin groups, problems would arise due to the changes in kin group size through time. The alternative paths for the transfer of land rights between susus, other than matrilineal inheritance, add flexibility to the land tenure system which allows land use decisions to be made more on the basis of environmental parameters. Within the family structure there is also some flexibility in usufruct rights. While their parents are living, both a husband and 90 wife have the choice of gardening on either fathers' susu's land, or either mothers' susu's. On either father's death, their rights to his land are extinguished. Each will maintain the rights to their respective mother's susu's land beyond her death. But while both parents and affines are living individuals have a wider choice of gardening sites in order to find the most fertile land within environmental constraints. If a couple are gardening on the husband's land but the available land under sufficient fallow is scarce, then they could choose to garden on the wife's land together. How often this flexibilty is used to avoid the less fertile gardening sites, or to leave longer fallows in order to not deplete gardening soils, is undetermined. Forest management decisions are quite often based on a strategy of gaining or maintaining land or plant rights. Ambiguity in the interpretation of land claims most often arises over unused land. Clearing and gardening reconfirms the rights. Maintenance of rights by showing use of the land is a particularly important issue now, when more landowners are way from their susu land for periods of time, by marrying or working off island. Gravesites and abandoned hamlets reduce the arable land base of a susu's land. But ecologically they may serve another purpose than agricultural production. In heavily gardened areas, they represent the only patches of regrowth to reach a late successional or mature forest phase. Whether or not primary species will recolonize such a site may depend on its microenvironment and its distance from primary forest stands. Their ecological role in the landscape is theoretical at this point, but these patches may be important in the dispersal and maintenance of primary forest species in a matrix of modified landscape. Depending on the type of plant community, land rights or plant rights will be more or less the issue of dispute. In sago groves, coconut and betelnut groves, and village agroforests the privately owned perennial crops are the prominent species in number and value and may or may not coincide with land rights. Yam garden land, with annual crops, is more subject to land dispute. In a shift toward a cash economy, the tendency at Nade will be to plant more land with groves of perennial cash-crops. The life-cycles of such crops are long, and the relation of individuals or groups with the land they are grown on is longer than under swidden cultivation. The transition from continuous plant tenure to the idea of individual ownership of land is not difficult, particularly when associated with the potent ideology of a market economy. If this is the trend, Nade concepts of land relations could be expected to become less flexible, and more formalized than they are now. The most appropriate land for cash-crops is the flat, easily accessible land near the coast. The coastal land is limited in area and not all lineages have claims there. While using the coastal land in this way is sensible in terms of the cash economy, there could be far-reaching implications for the total agroforestry landscape. It would have the effect of removing good land from the normal swidden-fallow cycle, which may lower subsistence efficiency and increase dependence on a market economy. 92 Chapter 9. FOREST RESOURCE MANAGMENT ON FERGUSSON ISLAND:  OPPORTUNITIES AND CONSTRAINTS OF THE TRADITIONAL AGROFORESTRY SYSTEM The traditional agroforestry system practiced at Nade appears to be common to most of Fergusson Island, although a few areas differ in the ratio of grasslands, sago, and other forest types that are managed. The system as practiced at Nade appears to have a role in the conservation of biodiversity, while at the same time sustaining sufficient agricultural productivity for the present population. However, there are external and internal pressures for cash crop and commercial forest development on Fergusson Island. Adapting the traditional agroforestry system may be one way to sustain agricultural productivity and alleviate pressure for the conversion of forest land into monocrop farmland, forest plantations, or grasslands. This preliminary study of the Nade agroforestry system provides some data useful to a land use planning procedure. More importantly it is the basis from which more questions need to be asked - by Fergusson Island people and land use planners, from both scientists and rural Melanesians with experience in development. This chapter sets the context for land use planning on Fergusson Island. Some possible land use options and suggested research priorities are laid out based on the above study and lessons that might be learned from the experience of development elsewhere in Papua New Guinea. 9.1 The development context in Papua New Guinea. 9.1.1 Forest development in Papua New Guinea. Throughout Papua New Guinea, increasing commmunity expectations and a national drive for development funding and foreign exchange, together place a strain on forest resources (Lamb 1977). Pressure for development comes from national political sources as well as local landowners. In many rural areas there is enthusiasm for development, and subsequent disappointment with the results. It has often been difficult to explain to landowners all of the options which are open to them and the consequences arising from each (Lamb 1977). Rural people are usually unaware of the benefits or risks to their surroundings and livelihoods from modern technologies. Objective scientific data on the impacts of commercial resource exploitation are scarce, and their implications are seldom translated into lay language for local people. Few rural people have ever seen land transformation on a massive scale. The agreements are often vague about costs and benefits except for the initial purchase price for trees (De'Ath 1980). Faced with a high powered logging or mining company eager to begin operations and a government eager for foreign investment, discussions with landowners are usually insufficient and ineffective leading to dissatisfied and reactionary responses (Viner 1984). Forests are one of the most easily exploited natural resources and can be seen as means for financing national development and ensuring national cohesion (Lamb 1977). At a regional scale, the rationale for encouraging logging and mining operations has been to finance roads into remote areas so that social development can take place. The forests of Papua New Guinea yield relatively poor commercial value in timber (Lamb 1977) because the species- diverse stands produce low volumes of marketable timber in relation to the cost of operations (White 1977). Once harvesting begins, there can be pressure to maximize extraction and neglect environmental and social concerns in order to assure profit (Ingram 1989). Lowland coastal rainforest and islands are particularly attractive for logging because of relatively inexpensive marine transport to processing centres. The most detailed studies of the social and environmental impacts of commercial logging in Papua New Guinea have been conducted for the large Gogol valley project near Madang. Seddon (1984), Waiko (1982), and De'Ath (1982) have outlined a host of negative social impacts arising from the virtual elimination of the resource base of the local subsistence economy, and the 94 tensions inherent in the conflicting value systems of traditional and introduced development processes (Hill and Hughes 1988). Forest conversion modifies not only the available resources but also the forest dwellers' social relations - by changing work patterns, demography patterns, introducing new kinds of disputes and gaps in perception between the younger and older people (De'Ath 1990). Saulei (1984) observed that there was substantial regrowth in the logged areas but that the successional vegetation was less diverse. He predicted long-term impacts such as water- logging, soil erosion and leaching, reduction of phosphorous levels, changes in floristic composition, and establishment of grassland in formerly forested sites. Despite the record of poorly planned, and socially and ecologically tenuous resource exploitation, Papua New Guinea would seem to have relatively favourable political conditions for an "ecodevelopment" style of forest resource planning. Ecodevelopment emphasizes reliance on local resources, autonomous goal setting at the local level, adaptation and generation of technologies for particular environmental conditions, and decentralized control over the style and rate of development (UNEP 1976). The social priorities which underly the concept of development must go hand in hand with land use which is ecologically sustainable (Ingram 1989). The population of Papua New Guinea is geographically isolated, and undergoing rapid social change, making local people very vulnerable to exploitation. At the same time the PNG constitution does recognize 97% of the land area to be under customary land tenure. And the constitution declares the fourth goal to be: for Papua New Guinea's natural resources and environment to be conserved and used for the collective benefit of us all, and be replenished for the benefit of future generations (The Government of Papua New Guinea 1975). The government is held together mostly by party coalitions, dominated by regional groups, rather than by political philosophy (Viner 1984). Thus it is sometimes possible for a small local group of people to make their views heard in the central government. Given these conditions, major development projects can be delayed or cancelled by a small but persistent community. 95 At the same time, the monetization of the rural economy of Papua New Guinea is inevitable (Pernetta and Hill 1983). Yet the importance of the subsistence component of rural economies should not be underestimated. Many material needs are still being met by products from the immediate environment. Throughout PNG, the forest provides food, medicines, ornaments, utensils, weapons, housing , and boats. The subsistence sector can meet the broad range of people's material needs more economically than imported goods, and the range of subsistence products can buffer the rural economy in times of crisis such as crop failure, and unemployment. To maintain a viable subsistence sector, or to introduce cash into a stable rural economy, both the physical resources and the knowledge and skills to efficiently use the resource must be conserved (Bulmer 1982; Clarke 1991). In a development process, adaptation of traditional agroforestry systems that provide a diversity of products can lower the risk of social and financial failure as well as ecological deterioration. 9.1.2 Customary laws and land tenure. Ownership is the central issue in both the development and conservation of natural resources in Papua New Guinea. Ninety-seven percent of the land and water, with the associated resources, is held under customary tenure by tribal and clan groups (Lakau 1988). For local people, their relationship with land goes beyond Western concepts of ownership to include spiritual and social dimensions. For most Melanesians customary land tenure forms the basis of social relationships. The land or territory of a group gives it social identity and cohesion (Eaton 1985). Speaking of Walfolo on the Northwest coast of Fergusson Island, Young (1987) says Land is really all the people have, and everything they hope to have will come ultimately from their land. It is their identity as well as their livelihood, for they know who they are by reference to their land (Young 1987:3). Altering land rights can arouse the same type of reaction as interfering with religion (Dahl 1984b). Land is held in such a complex system of rights and obligations that transfer 96 of land to individuals or groups outside these relationships is rare. Land uses like plantation forestry or conservation are particularly difficult to implement under customary land laws because they require a long term committment to a specific use and may show no immediate and obvious benefits to landowners. People may be willing to sell rights to harvest timber, but not to commit land to a future use (Eaton 1985). This has actually prevented reforestation in many areas. The communal nature of ownership, fragmented holdings, multiple rights-holders and lack of any clear title make it difficult for entrepeneurs or the government to acquire land for the purpose of development. Even when land is acquired the disputes and claims for compensation are ongoing. In some cases the proper payment is not made, or agreements are misinterpreted. Melanesians do not ever really mean to sell land in the Western sense of a sale (Lakau 1988). They cannot ever fully dissociate themselves from their land. Customary land tenure systems have been successful in maintaining a fair allocation and sustainable management of scarce resources. Traditional conservation practices are closely linked to customary land tenure system. Group controls prevent the alienation of land outside the group and limit dispossession and unequal relationships (Eaton 1988). Access to resources is limited to those with traditional rights. Taboos with a basis in religious belief and magic prevent over-exploitation of hunting territories. Some resources are sacred, such as bird of paradise display trees, bush fowl nesting grounds, or sacred areas considered the home of the spirits. Swidden cultivators have a keen sense of the value of the forest to them for its ability to regenerate soil fertility, and when possible adjust the swidden cycle accordingly. Most of these practices are not consciously motivated by conservation objectives or the conscious management of resources for sustained yield (Bulmer 1982). The effectiveness of these traditional conservation practices are sometimes romanticized and exaggerated. In a pragmatic way people's concern is often with relatively short term yields. Early inhabitants of Melanesian islands did not practice a conservation ethic preserving an unchanging paradise until Europeans brought major disturbances; instead they caused many 97 extinctions, reduced forest cover, initiated massive soil erosion, created and extended grasslands (Clarke 1991). Nonetheless, in transforming natural landscapes into cultural landscapes, the early inhabitants developed sustained-yield systems of agriculture and agroforestry that are still producing today. Throughout Papua New Guinea customary law and land tenure help to maintain that balance between rural societies and the resources they need (Eaton 1988). But the balance is more and more precarious. The cash economy, population increases, new technologies and changes in the traditional social structure have created new needs and pressures on resources. There is an increasingly individualistic approach to land use and less communal controls. The focus is often on compensation, royalties and quick profits (Eaton 1988). Building on customary systems of land tenure and management, and ensuring that there is obvious benefit to local people is the most effective approach to both resource development and conservation. Unfortunately, the establishment of protected areas has sometimes destroyed traditional patterns of tenure in the name of conservation (Ingram 1989). 9.1.3 Conservation area planning. Throughout the tropics traditional land use systems may provide the basis from which a range of alternative land uses and conservation strategies can be developed without destroying the viability of existing social and natural systems (Yen 1977). Both conservationists and development workers are aware of the possibility, indeed the necessity, of integrating rural development and conservation objectives. Unless people have a direct stake and interest in conservation, then the most well-designed projects stand little chance for long-term success. Fundamental to the conservation and sustained use of the forest is to show that standing living forests are more valuable than cut burned ones (Posey 1989). Certain agroforestry activities, in buffer zones surrounding core reserve areas, can function to protect the reserve from encroachment and negative impacts from fringing 98 cultural or settled lands (Von May dell 1991). And traditionally managed forest plots and agroforestry systems can contribute to the success of forest reserves by providing an alternate source from which local people can harvest useful plants (Alcorn 1981). In incorporating or adapting traditional systems into rural development and conservation of biodiversity, the use of local knowledge of the local species and ecosystems is essential for applying ecological principles in site-specific and culturally appropriate ways. The ecological processes, found to be typical in polyphase and polycultural agroforestry systems, may contribute to long-term ecological sustainability in rural areas, and conservation of biodiversity. These processes include: high rates of biomass accumulation, closed nutrient cycling and biological control mechanisms for weeds, pests, and disease (Gleissman et al. 1981). The micro-climate of a closed canopy agroforestry system is intermediate between open field and closed climax forest. This would affect the movement and establishment of plant and animal species at the boundary. Forest patches integrated into agroecosystems on a regional or national scale provide additional habitat for many wild species (IUCN 1980). Traditional agroforestry may have a role in the design and management of networks of protected areas. Agroforestry is inappropriate in the core zone of primary forest conservation areas. But agroforested buffer zones can contribute to the stability of forest reserves by allowing environmental "edge-related changes to take place without affecting the large stable core areas" (Lovejoy et al. 1986). Polycultural agroforestry systems that maintain a high species and genetic diversity serve as a valuable in situ gene bank for wild forest species as well as land races and wild relatives of food crops (Alcorn 1984). Biological evolution is a dynamic process that includes species extinctions, but recent human interventions in ecosystems around the world are accelerating the reduction of species and genetic diversity. Valuable crop genetic resources are threatened by habitat disturbance. Areas are cleared of natural diversity and planted with exotic varieties (National Academy of Sciences 1989). High yielding uniform 99 cultivars of a few crops are replacing the genetically variable indigenous varieties commonly grown by subsistence farmers (Cohen et al.1991). Planting large areas of genetically uniform cultivars makes crops more vulnerable to unexpected events of drought, frost, pests, and blight epidemics. Trees selected and bred for use in plantation forestry often have genetically simpler resistances. The result is a reduction in the genetic base and a decrease in the genetic flexibility to meet unforeseen environmental change (Thelen 1984). Moreover, when the new commercial plant varieties are grown in traditional agricultural systems they tend to be less dependable than the varieties they replaced (Altieri and Merrick 1987). In situ refers to the maintenance of genetic resources in natural settings (Brush 1991). In situ conservation allows for continued dynamic adaptation of plants to the environment (Ingram and Williams 1984). By preserving habitats, germplasm is maintained as well as the processes that create new germplasm. For crop plants this is especially significant. In situ genebanks function as living laboratories. Species maintained in their natural habitat allow for ongoing study of their ecology; an opportunity not available in ex situ genebanks (Prescott-Allen 1982). The traditional practice of polyvariety cultivation offers a practical means to preserve endangered germplasm resources of the world's food crops (Clawson 1985). Under multicrop agriculture, crops can be enriched by gene exhange with wild or weedy relatives (DeWet and Harlan 1975). Weedy crop relatives thrive in disturbed areas, and often live on the margins of fields or fallowed areas. Land use practices such as clearing, weeding and fallowing can imply in situ conservation for weeds (Brush 1991). Genetic resources of commercial tree species are preserved in some forest reserves, but otherwise forest management is not done with genetic conservation as an objective. Tree germplasm is used in its primitive forms and depends less on human agency than weeds or domestic crops. In situ conservation of tree species also requires special attention to the population biology of the target species (Brush 1991). Nonetheless, some traditional agroforestry systems may 100 provide the appropriate environments for the conservation of commercial tree crop germplasm. While traditional cultures are responsible for maintaining important repositories of crop germplasm (Alcorn 1984), it is also true that "artificial attempts to halt social change and sustain or revive traditional cultures cannot achieve conservation" (Bulmer 1982:76). Allowing local people to continue traditional practices in or around protected areas can make them allies in the conservation objectives as well as preserve valuable empirical knowledge. This does not mean preserving indigenous cultures as museum pieces. If and when local people see the opportunity for economic stability in the context of ecological sustainability then they may be interested in protecting a piece of wild environment (Dasman 1982). 9.2 The potential for ecodevelopment on Fergusson Island. The people of Nade would like to be more active players in a cash economy. They talk of wanting development for Nade, but have no clear conception of what kind of development nor specifics of the process. Nade people's attitude and their history of unsuccessful attempts at incorporating cash crops is probably typical of Fergusson Island communities. The island communities of the D'Entrecasteaux Islands in general have few short-term prospects for significant growth toward a cash economy. There are virtually no local markets for agricultural products and no means to transport cash-crops to towns or ports. While many people are content living a subsistence lifestyle, others are not and may increasingly choose to emigrate to find jobs in the towns; becoming the major cash generating export of the islands. Nade people are aware of the results of the Ulabo logging operation on the east coast of Fergusson Island, and most claim to not want such foreign controlled development near their village. The amount of merchantable timber left on Fergusson Island is said to be insufficient to attract a large commercial operation at this time (William Wapiti, Esa'Ala district forest officer, personal communication). A small sawmill is presently operating in 101 the Ulabo area and could be viable once the technical difficulties are worked out. In short, some forest management options for Fergusson Island involve conservation planning, developing ecotourism, allowing selective forest cutting for small sawmills, the conversion of forest to cash crop plantations such as cocoa, coffee, or oil palm, and/or continuing traditional swidden and agroforestry practices. None of these options are mutually exclusive. 9.2.1 Conservation and ecotourism planning Within the Pacific region, Fergusson Island has been identified as an important focus for conservation because of its ecological richness, complex mosaic of ecosystems, and large tracts of primary rainforest. Beehler (1985) proposed a reserve on Fergusson Island to form part of a network for the protection of rare New Guinea birds, and to protect in particular, the unique Goldie's Bird of Paradise CParadisea decora), the curl-crested manucode (Manucodia  comrii) and the trumpetbird (Phonygammus keraudrenii). A small Wildlife Management area, proposed and managed by local people, was established at Lake Lavu, in the interior of Fergusson Island, to protect crocodiles living in and around the lake. There are bans on firearms and hunting by outsiders. However, logging began on the east coast of Fergusson Island in 1987 without any acknowledgment of the conservation proposals and without a management plan. Without a grassroots effort to develop a land use plan for the island, some key habitat types and locally occurring species will probably disappear (Ingram 1990). Scientific knowledge of the diversity of species and habitats on Fergusson Island is lacking. To begin monitoring biodiversity and the sustainability of land use practices local ecological knowledge should be utilized. Site-specific ecological knowledge is highly variable between and within communities of Fergusson Island (Ingram 1989), and much of the knowledge is held within lineages with secrecy and suspicion. Knowledge can be a form of power, particularly in relations between young and old. But, it might be that an exchange of 102 information amongst communities will facilitate the political cohesion necessary for a grassroots, pro-active land use plan. The low human populations, a viable subsistence economy that relies on a range of wild species, and local land ownership on Fergusson Island would seem to provide an ideal situation for the establishment of a conservation area compatible with the needs of local residents (Ingram 1989). However, a proposed conservation area would require a cooperative committment of land by several clans within Nade and surrounding villages. There is littie political cohesion within villages or between language groups on the island. There is no one traditional leader for any village. The elected village councillors provide little focus. Church structures may represent the only unifying institution. Within the Massim group, there is suspicion and mistrust between neighbours that is rooted in the tradition of sorcery. An individualistic approach to land use inhibits the communal control needed for applying internal restrictions to hunting and gardening. For purposes of conservation of biodiversity, the most important aspect of the Nade traditional agroforestry system is the maintenance of a diversity of successional phases and a wide range of crops species and varieties. Many plants currently being managed at Nade are potential genetic resources of national and international importance; for instance, wild relatives of banana, Musa spp.; Pandanus spp.; wild mango, (Mangifera spp.); wild sugar cane, Saccharum spp.; wild taro; and wild citrus (Ingram 1989). The genetic resources of the staple crops of Fergusson Island are also important. The New Guinea region is a centre of diversity for many root and vegetable crops such as sweet potato, yams, taro, and aibika. It is the centre of origin for sugar-cane, bananas, and sago (Powell 1982). However, the people of Nade seem to have no conception of the loss of biological diversity, or of the long term affects on the forest of current subsistence practices, and no conscious awareness of their impact on forest as a group. They still perceive Nade as surrounded by uwama, or big forest, and individuals seem to have no reluctance about cutting further into the primary forest if they so desire. Land use decisions are made within 103 the susu, and do not consider Nade as a whole. Fortunately, more arable land is not immediately needed, the labour required to cut primary forest is inhibitory, and land tenure constraints are not yet forcing the expansion of agricultural lands. The people of Nade perceive there to be "plenty" of Goldie's Bird of Paradise and they see little need for regulation. Shotguns were banned in 1988. In any case, ammunition is prohibitively expensive for most villagers and obtaining it involves a two day trip to the provincial capital of Alotau. While there is no hesitation for people to shoot a bird of paradise with a rubbergun (they see it as their right), there is rarely a reason to. The feathers are used as decoration in the occassional dances and are prestigious in that context. The birds are respected as beautiful, intelligent, and difficult to shoot. They are said to "talk to peole and to dance to people's voices" (Tony Dorsy, personal communication). Traditional conservation practices exist, but are more incidental than for the objective of long-term sustainable resource use. Outsiders are excluded from hunting or fishing in Nade territory and clans restrict each other from clearing gardens in some areas. The use of plants and wildlife may be restricted to certain individuals thereby making over-exploitation less likely. Sacred sites are left undisturbed. These practices and restrictions help to maintain somewhat of a balance between Nade society and its resources. But the motivation of these practices is not nature conservation. As new needs and technologies are introduced this balance will be more and more disturbed. The Goldie's Bird of Paradise is known to have attracted the occassional Australian ecotourist group, and Lecroy, Ingram and myself as researchers. This also gives the bird some value to the people of Nade. Indeed, the development of small-scale ecotourism, if carefully planned, could encourage local conservation awareness. Fergusson Island people have a rich system of knowledge and belief about the environment and this could be integrated into a program of education which could introduce modern concepts of the ecosystem and its functioning. But a conservation program must do more by providing short-term benefits. 104 The monetization of the rural economy of Fergusson Island is in process. Problems may occur if people feel that the establishment of a conservation area prevents them from enjoying the benefits of economic development, such as timber rights, purchase fees, and royalties, as enjoyed by their neighbours. The success of establishing a conservation area will depend on the participation of local people. Only short-term yields will convince local people to participate. Conservation has to be made to pay and seen to be paying (Buhner 1982). The value of the Goldie's Bird of Paradise and its habitat to the people of Nade will need to become more explicit. Benefits to local people could be derived from development of buffer zone agriculture or ecotourist enterprises. The harvesting and selling of high value forest plant and animal products could be done on a limited basis, along with forest species enhancement. Products might include: fungi, medicinal plants, wild orchids, butterflies and species of scientific interest. 9.2.2 Adapting traditional agroforestry. The agroforestry system at Nade could be an appropriate activity for a buffer zone to a forest reserve area for several reasons. The system allows some cultivated areas to reach a mature secondary forest phase. Some patches of primary forest are left undisturbed. The mature fallow phase provides wildlife habitat. The introduction of exotic plant species is continuous, but on a small-scale. No pesticides or fertilizers are used. An almost continuous soil cover and mixed cropping provide a transitional microenvironment to mature forest. The majority of subsistence products are obtained from fallow phases and swidden gardens. Cultivation intensity is currently at the limit of being compatible with conservation objectives. In certain locations, the fallows have been shortened. Further population increase, plus the introduction of cash crops could intensify the system to the point of degrading the environment and decreasing levels of biodiversity. The agroforestry system could be adapted to ensure sustainability and eliminate the need for encroachment into a conservation area. Incorporating the appropriate cash crops into the agroforestry system 105 would be one way to make conservation pay. Changes might include green manuring, or the use of fast-growing leguminous tree species as shade crops in swiddens, or to improve fallows. The soil could then sustain a shorter fallow cycle, or a longer cropping season. A major concern in agroforestry research is sustainability. Sustainability in a culturally modified landscape is determined by the ecological structure and function of the system as well as its continued ability to fulfill the socio-economic needs of the people (Soemarwoto 1987). To maintain sustainability, any improvement of a traditional agroecosystem must be grounded in ecological principles and be economically viable. The promise of agroforestry is that trees can serve to plug many of the holes in tropical farming systems. Selecting the appropriate tree species and how many should be added to or retained in the existing system depends on the useful niches for trees. An agroforestry niche has both a functional role in the land-use system and a place in the landscape (Raintree 1986). There are numerous examples from around the Pacific, of combinations of annuals and perennials that transform shifting cultivation into more permanent agroforestry practices. In Samoa, Erythrina intercropped with taro and yams. In PNG, Leuceana planted with sweet potato, and in the Solomons Leucaena spp. and Gliricidia spp. shade cacao (Vergara and Nair 1985). Casuarina oligodon is known to fix nitrogen and is good source of timber. In the PNG highlands it is planted on the edges of gardens and in fallows to produce "controlled fallow" (Wood and Humphreys 1983). In the Trobriands and Woodlark Island, Schlienitzia sp. is cultivated as yam posts, adding what locals term "grease" to the soil (Neville Howcroft, personal communication). Budelman (1990) investigated the use of three leguminous trees as live support systems in yam cultivation; Leuceana leucocephala. Flemingia macrophylla, and Gliricidia  sepium. Gliricidia sepium was the only one of the three to give a high yam tuber yield. The appropriateness of a tree species as yam posts was found to be a function of specific properties: low leaf productivity, relatively weakly developed root system and an open 1 0 6 architecture - properties that leave enough space above and below the ground for a yam crop to grow in association. The advantages of using local species in a new cropping system is that their biology and ecology is already familiar to local people. This may reduce the perception of risk by local people and encourage their participation in the design and troubleshooting maintenance of the adapted cropping system. Using local species instead of introducing exotics also increases the ecological integration of the system into other successional phases. And local species, endemic to Fergusson Island may represent genetic resources which are valuable and worth retaining, in their own right. There are local tree species on Fergusson Island that, although untried in agroforestry in Milne Bay province, have obvious potential. Casuarina equisetifolia occurs on the coast of Fergusson Island and may prove to be a good timber species and a nitrogen-fixer. The leguminous trees Pterocarpus indicus. and Albizzia falcatoria are abundant in the Nade landscape and trials may show them to be appropriate agroforestry species. Erythrina  variegata is also present, and has been used in agroforestry systems elsewhere. If the addition of cash crops into the Fergusson Island economy is desirable, then the possibility of integrating them into the existing diversified agroforestry system could be explored. The alternative is to introduce commercial monocropping which requires more intensive land use and modification and encourages the expansion of agricultural lands into primary rainforest. The use of diverse and local species in an agroforestry system lowers the risk of genetic erosion and a diversity of products lowers the risk of economic failure due to market fluctuations or crop failure. Adaptation of traditional systems would facilitate the transition from a subsistence to semi-subsistence or cash economy. While there is potential for commercializing traditional agroforestry systems by adding cash crops, there are many risks associated with it. Commercialization often involves decreasing the diversity of species and/or varieties and increasing the use of exotics, which may increase the process of genetic erosion (Soemarwoto 1987). In commercial agroforests 107 with low species diversity, harvesting is seasonal instead of continuous. Such an agroforest no longer functions as a living store from which owners can harvest according to the nature and time of need. The control of species composition and harvest changes from being internal, by the gardeners themselves, to external, by the market forces. Additions of high-yielding cash crops usually represent higher energy input and higher risk. Pests and diseases, market-price fluctuations and high seasonality reduce stability (Conway 1987). Changes in labour roles or the tenure system, to accomodate the new cropping system, often creates social stresses. Productivity may be increased, but at the expense of stability, equitability and sustainability. The potential for creating transportation, storage and processing facilities for cash crops harvested at Nade would need to be explored. An extension program to facilitate the introduction of a new cropping system would also be required. The most obvious cash crop addition to the Nade landscape would be cocoa. Storage, processing and marketing would be facilitated by the parallel development of cocoa cropping already being organized by landowners in the Ulabo timber area (20 km away on Fergusson Island, and within the Esa'ala administrative district). Transportation of the cocoa to a market would need to be organized cooperatively within the Nade community. Cocoa is already the basis of a simple agroforestry system used throughout Papua New Guinea with Gliricidia sepium interplanted as a shade crop. Gliricidia is proven effective in most areas, though it can be susceptible to pests. It could be readily obtained from the district forest office at EsaAla. All the social and ecological implications of adding cash crops to the traditional subsistence systems of Fergusson Island cannot be foreseen. But some trends and responses might be projected based on my observations at Nade, and lessons from resource developments elsewhere in Papua New Guinea. The adoption of agroforestry practices is influenced by socioeconomic and cultural factors as well as biophysical considerations 108 (Vergara and Nair 1985). Successful introduction of cash crops into the landscape will depend largely on land tenure issues. The distinction between rights to plants and rights to land must be recognized in planning for the commercialization of agroforestry (Weinstock and Vergara 1987). As cash crops give new value to land, the traditional system for the transfer of land is disrupted. Often, when the value of trees is increased there is a tendency of both land and tree tenure to shift from communal to private (Fortman 1985). While rights to specific plots of garden land are temporary and shared by clans, ownership of plants is more permanent and private. If trees and other perennial crops are cultivated, individuals secure rights that will last for the life of the crop. The clan retains ownership of the land, but other members are precluded from using it. Thus the perennial components of agroforestry would not be encouraged by clan leaders at Nade because they can lead to extended control by the planters. Planting trees may be used as a means to get or maintain rights to land. People keep no written records, relying on oral transmission of tenure arrangements to younger generations. Forgetting the terms of an extended use is a complication that landowners would try to avoid (Weinstock and Vergara 1987). A solution might be to develop clan-controlled rather than individual household-owned agroforestry farms (Vergara and Nair 1985). When land is planted with cocoa, coconuts, or other perennial cash crops, it is put out of circulation for some time, reducing the flexibility of the tenure system to respond to demographic changes. In the Nade landscape, a cash crop agroforestry system would represent one of the more permanent zones within the life-cycle of the shifting forest mosaic. It would most likely occur in patches integrated with the existing successional phases. The most suitable soils are in the Ogauga area of Eastern Nade. This area is limited in extent and already under land dispute. Introduction of a cash crop would likely exacerbate the situation. Developing a niche for cash crops in a conservation area buffer zone would be one way of making conservation pay. But more ecological research would be needed to develop 109 a productive and sustainable agroforestry system that also conserves biodiversity. An inventory of biodiversity and a framework for ongoing monitoring could be established. An extension program and market development would be needed to ensure financial success. Local groups at the susu, village, or church level would need to open communications and begin to organize the application or the adaptation of the customary land tenure system to a semi-market economy. 110 Chapter 10 CONCLUSION This study produced a baseline inventory of landscape types and of culturally and ecologically significant plant species for Nade ward, Fergusson Island. The inventory can be used to monitor changes in land use patterns and biodiversity on Fergusson Island, and to facilitate a future land use planning process. The identification and mapping of the mosaic of plant communities revealled a complex polyphase agroforestry system. The vegetation phases are ecologically integrated in both time and space. Spatial distribution of the phases depends on local environmental conditions which vary within the subsistence territory of Nade. But the mosaic is far from static. Overlying the environmentally determined pattern of the shifting mosaic are the social factors. Land and plant tenure, labour-saving strategies, and agricultural traditions affect land use decision-making, in turn affecting vegetation patterns. This study has shown that the subsistence sector of the Nade economy is vital. The various phases of the agroforestry system provide a diversity of foods and at different times of the year. The system also meets material needs such as fuel, housing, boats, tools, utensils, ornamentation, weapons, and medicines. The plant resources of the Nade landscape provide more than the minimum needs for survival. They also can support recreation, sociability, artistic creativity, and spiritual contemplation. A subsistence lifestyle should not be romanticized. Nade people do show a desire to participate in a cash economy. The integration of cash crops into the existing agroforestry system, and the development of ecotourism, or a market for small-scale crafts from forest products are some possibilities. Fergusson Island has been identified as a priority in Oceania for conservation of biodiversity. Further research on the ecology of the traditional agroforestry practices should confirm the compatibility of those practices with objectives of habitat conservation. However, local management of an agroforested buffer zone to a conservation area would - . I l l require unprecedented cooperation between local land-owning groups. And to generate local initiative, conservation must be seen to pay. From some traditional agroforestry systems in tropical rainforest we have an opportunity to learn time-tested ways to maintain forest resources, maintain (or increase) food production, and to lessen dependence on expensive imported agricultural inputs. If the wealth of local traditional knowledge of the environment is to be used for conservation and development purposes, it has to be recorded, evaluated and integrated into the scientific record by people with an interdisciplinary background in the social and environmental sciences. Most islands of tropical rainforest are inhabited by rural people. Both forest conservation and development planning will be facilitated by a respect and critical appreciation for local expertise. In times of social and technological change, the use of local knowledge and the adaptation of traditional land use practices can help to empower local communities in their own self-development. 112 FOOTNOTES 1. There has been debate over the use of the term shifting or swidden cultivation (Conklin 1957; Lea 1964). Here swidden cultivation is used to refer to the agricultural practice of clearing forest, planting crops, leaving the site to fallow and returning to the site 7-20 years later. Conklin (1961) discusses the history of this agricultural practice in its various forms -in primary or secondary forest or grasslands, crop-fallow time ratios, types of crops, dispersal relative to human settlement, and specific tools and techniques used. 2. Allen and Crittendon (1987) suggest that the spread of horticultural techniques for sweet potato such as tilled beds and compost mounds was related to the spread of the moka and tee exchange systems in the highlands. However, the exchange systems were inflationary and competitive, so the techniques used in restricted environments were spread to less suitable environments where they have caused considerable land degradation. 3. Greg and Ema Nimagole. Neither Greg nor Ema are from Nade; Greg is from Budoya, Fergusson Island, and Ema from Kiriwina, Trobriand Islands. 4. The people of Nade were both shy and curious in my presence at the beginning. They were embarrassed at their bush houses and bush food. They claimed to be unaccustomed to housing guests and told me this is because they do not participate in Kula and so never learned customs for welcoming guests. But the formality with guests appears to have been evident in Dobu too. According to Fortunes account "Visitors are never invited to share a meal. They may be given food, but if so they eat it apart with their backs turned to the givers.... Visitors never enter the house. They sit underneath the house or on its projecting front verandah. Visitor's also never enter another person's garden." (Fortune 1932:75). I was welcomed into gardens by the owners but was warned not to go to gardens alone. 5. Ema was known for her sociable nature, and for always having and sharing tobacco and betelnut. I became an added attraction for those people who would regularly come around in the evenings to tell stories, discuss the days events, and to question or be questionned by the dimdim (white foreigner). 6. One of the most studied aspects of Massim culture is the Kula trade ring (Leach and Leach 1983; Young 1971). The Kula is an elaborate exchange system extending from the Trobriand Islands in the north to Woodlark Island in the east to Tubetube Island in the South, and Fergusson Island in the west. Of all the regional exchange systems in Melanesia, the Kula is of interest as one of the few to survive missionization, colonization, and the advent of a cash economy (Macintyre and Young 1983). » Anthropologists have often viewed the Kula as primarily a ritual or gift exchange system, however prehistorians have challenged that view by trying to reconstruct the trading cycles according to geographic and ecological models (Lauer 1970, Egloff 1978). Trade relations may have been formed to gain access to scarce or exotic resources. The islands of the region vary greatly in geology and soils, and thus vary in the suitability for the production of specific goods, and the surplus needed for trade. Some islands lack clay deposits for pottery. Murua Island was the only source of a particular stone suitable for stone axes. Fergusson, a volcanic island, was the only source of obsidian. Some islands hosted tree species especially suitable for canoe carving. The most fertile islands allowed for a surplus of yams and other foods, while others were stoney or infertile, or more subject to drought (Macintyre and Young 1983). The subsequent economic interdependence may provide a rationale for the establishment of the trading system,'but it cannot explain the cultural processes by which such an elaborate system evolved (Macintyre and Young 1983). Further, it cannot explain the 113 persistence of the ceremonial exchange even though the traders participate more and more in the cash economy. Yet, regardless of the debated why and how of the Kula, its significance to the context of this study is that it would seem to have been an obvious avenue for the transfer of horticultural techniques within the region. Also, such an extensive exchange system may have regulated subsistence patterns by encouraging the production of surplus of some goods. The people of Nade have never participated in the Kula trade circle but they are well aware of its functioning. Many villagers have relatives from active Kula villages such as Dobu, and some people in Nade have acquired old shell necklaces and armlets through relatives. This may be some indication of the degree of Dobuan acculturation in Nade. 7. Indigenous enterprises are on a very small scale. Accumulation of native wealth and the manipulation of capital is not part of the local political system. Sorcery is feared by those who might display exceptional talent or conspicuous wealth, whether it is counted in gardens, pigs, shell valuables or cash. "The D'Entrecasteaux islanders as a whole affect an uncompromising egalitarianism. These are probably important factors in the weak personal incentives of would-be entrepeneurs." Young 1971:10. 8. Ulabo timber is a New Zealand owned logging company with the licence to log an area of lowland forest (up to 600m) on the east coast of Fergusson Island. Logging was in operation from 1987-1989. Most of the area was selectively cut. The major genera taken out were Dillenia (approximately 75% of the harvest), Ficus, Terminalia, Canarium, Planchonella, Pterocarpus (rosewood), Alstonia. Xanthomyrtus. The logs were being shipped to Japan for veneer. The local people were expecting royalties, and development of roads and cash crops as compensation for their land. The royalties have been slowly received. People claim they have not received as much they expected. The roads were made initially, but there is no equipment for upkeep. The cash crop development is starting (May 1990) to be organized by a liason officer with the Esa'ala Development Authority. At the time of this research a few farmers had planted cocoa with a cover crop of Gliricidia sepium. It is too soon to judge the success of the project. 9. I observed the relatively small feast prepared for the grade 6 graduation which involved weeks of planning and preparation. Yams and other garden food were put aside for months in advance, food was sent from relatives in towns. 10. Legend of the Budowai tree. A large budowai tree (Alstonia scholaris) grew over a river where the women bath. It was believed that when a leaf fell in the water a woman would get pregnant and give birth to an albino child. This happened to a woman. Her husband suspected another man and became angry. When it was explained he cut down the tree. 11. Fortune (1932) gave an account of the degree to which the fear of sorcery and witchcraft pervaded Dobu society 60 years ago. "Disease and modes of death that are indigenous are well known and catalogued. Their production and infliction upon near neighbours is one of the customary occupations of the people. The whole life of the people is strongly coloured by a thorough absence of trust in neighbours and the practice of treachery beneath a show of friendliness. Every person goes in fear of the secret war, and on frequent occasion the fear breaks through the surface." (Fortune 1932:137). Fortune claims that people recognized correctly the diseases of introduced origin such as measles, tuberculosis, influenza, and dysentry. According to Young (1971) the complex of sorcery beliefs on Goodenough are identical to those described by Fortune. Jenness and Ballantyne (1920) and Chowning (1969) describe similar sorcery beliefs. 114 12. An obstacle to learning how pervasive the belief and use of magic is in Nade today is the secrecy that surrounds it. "(the employment of) white magic is not harmful to the rest of the community. No man would be at pains to deny knowledge of this kind of magic; on the contrary he is proud of whatever little he may know, and is always eager to learn more. For it is plainly all to his advantage if he knows a charm to make the yams grow big, or to win a girl's affections, or again to heal the sick. But there is another kind of magic, black magic, which no one will confess he knows. Any native will tell you how it can be done, but he will always say that he himself is ignorant of the proper incantations." (Jenness and Ballantyne 1920). 13. In Nade, each susu has its own yam magic. "They say that only the seed descended within the susu will grow. If seed could ever be obtained as a gift from other susu, as it cannot, it would not grow anyway, far less seed from other villages....different villages of different totems have different garden magic. Yam seed is only handed down in the susu line, from a man to his sister's son." (Fortune 1932:70). 14. Betelnut chewing is a central part of Massim life. It is called magi in Nade. Chewing is a social activity. Adults chew anytime; when they sit with friends, work in the gardens, attend feasts, go to church, and during canoe voyages. Children start chewing at about 7 years old. In Dobu, betel nuts are placed in the hands of a corpse for the guardian of the place of the dead so he will let the spirit pass (Fortune 1932). Betel-nut (Areca catechu) is chewed with the fruit or leaf of the betel plant (Piper  betle. a climber) and mineral lime (burnt coral). "It appears that the euphoria is caused by the nicotine-like properties of the alkaloids in the betel-nut and that these alkaloids are released from the nut by the lime. Chewers report that the saliva is reddest and that the chew tastes best when all three ingredients are used together." (Beran 1988) 15. Though I was unable to observe the sago harvest, the process as it was described to me is the same as described by Jenness and Ballantyne (1920). "The tree is first felled, and its' outer sheath removed to form a trough. Near its' lower end a sieve of fibre is attached, and the pith of the sago is dug out and placed in the upper part of the trough. Water is poured over the pulpy mass, and it is firmly kneaded and squeezed against the seive, when the nutritious starch passes through in suspension and settles at the bottom of a bowl placed under the mouth of the trough. Then it is wrapped in leaves and hung to dry." (Jenness and Ballantyne 1920:31). 16. The namesake system is a means to strengthen ties of an outsider to a susu or to strengthen an "adoption" arrangement. When a child is named after someone then he or she becomes effectively part of the namesake's susu. The rights that go along with namesake-membership in a susu seem to vary with individual need. If the child is born into a matrilineal line with all the usual rights and obligations then the namesake relationship may only involve some food exchange. If the childs lineage is unclear he/she may be adopted into the namesake's susu with rights to land and the obligations of a full member. 17. In Nade, I saw signs written in Dobu language attached to betelnut trees to claim ownership. In Dobu (Fortune 1932), and on Fergusson Island (G.B.Ingram, personal communication) it is also common for ownership of trees to be protected by magic. " On coconut and betel palms away from the village, there is to be seen the customary dried coconut palm leaf tied round the trunk....the leaf is a warning that a spell for causing elephantiasis, gangosa, tertiary yaws or the like, has been placed upon the tree trunk. A thief will touch it at his peril....different susu of the same locality own the spells for different diseases. This ownership is hereditary within a susu line of descent, and one susu will never sell its peculiar powers to another." (Fortune 1932:80). 115 18. In Dobu, "a strictly kept rule is that a child may not eat of the fruit of dead father's trees. The abstinence of a parent towards his or her father's trees is enforced on a child in turn. Accordingly, if the same tabus were bequeathed by a man to his son and to his sister's son, the former could not use the tabus to pilfer the trees of the latter compatibly with keeping the rule of abstinence, nor his son after him. But the patrilineal line would have no comparable protection from the matrilineal line. In the case of tabus, the abstinence from the fruit of a dead father's or grandfather's trees is kept fairly rigidly and ensures exclusive susu inheritance." (Fortune 1932:82). 19. Again it is difficult to know how much magic is practiced at Nade. Certainly, the distrust and suspicion is evident. Fortune describes yam stealing in Dobu. "It is believed that yams can be enticed out of one man's garden into another man's by the latter's magic. Everyone, the most respectable included, practise magic to steal the crops of other persons' gardens in this way. It is done every year by all gardeners, in private, as a matter of course. The yams are believed to emerge from the gardens and walk about in the night. Then they can be enticed away from their home garden by magical incantations. Stealing here is believed to be done by magic alone. The person who steals in the flesh from another person's garden is not despised for stealing as such, but for stealing in a clumsy, inefficient, and ridiculously pride forswearing manner. It is base to admit such poverty, as is admitted by garden stealers in the flesh. It is degradation to eat into one's stock of seed yams, not setting aside seed enough for a reasonably good garden." (Fortune 1932:83). 116 L I T E R A T U R E C I T E D Alcorn, Janis 1981. Huastec noncrop resource management: implications for prehistoric rain forest management. Human Ecology 9(4):395-417. Alcorn, Janis 1984. Development policy, forests, and peasant farms, reflections on Huastec-managed forests: contributions to commercial production and resource conservation. Economic Botany 38(4):389-406. Alcorn, Janis 1989. Ethnobotanical knowledge systems: resource for meeting rural development goals. In: Indigenous knowledge systems: the cultural dimensions of  development. M.Warren, D.Brokensha, and L.J.Slikkerveer (eds.). London, UK, Kegan Paul. Allen, Bryant J. 1988. Environmental ethics and village agriculture. In: The ethics of  development: choices in development planning. Papers from the 17th Waigani seminar. Philip J. Hughes and Charmian Thirwall (eds.). pp.6-21. Port Moresby, University of Papua New Guinea Press. Allen, B.J. and Crittenden, R.G. 1987. Degradation and a pre-capitalist economy: the case of the Papua New Guinea highlands. In: Land degradation and Society P.Blaikie and H.C.Brookfield (eds.). London, Methuen. Altieri, M. 1990. The ecology and management of insect pests in traditional agroecosystems. In: Ethnobiology: implications and applications. Proceedings of the first international congress of ethnobiology. Vol.1 Darrell Posey et al (eds.). pp.131-145. Belem, Brazil. Altieri, M. and Merrick, L. 1987. In Situ conservation of crop genetic resources through maintenance of traditional farming systems. Economic Botany 41(2):86-96. Altieri, M. and Merrick, L. 1988 Agroecology and in situ conservation of native crop diversity in the third world. Biodiversity. E.O.Wilson (ed.). pp.361-370. Washington, D.C. National Academy Press. Atchison, M.B. 1974. Summary of Dobuan culture today. Mimeograph on file in New Guinea collection. University of Papua New Guinea, Port Moresby. Balgooey, J.Van 1976. Phytogeography In: New Guinea Vegetation K.Paijmans (ed.). pp.l-22. Canberra, Australia, Australian National University Press. Balick, M.J. 1990. Botany with a human face. Garden Nov/Dec 1990:2-3. Barker, John 1989. Western medicine and the continuity of belief: the Maisin of Collingwood Bay, Oro Province. In: A Continuing Trial of Treatment. S.Frankel and G.Lewis (eds.). pp.69-93. Dordrecht, Netherlands, Kluwer Academic Publishers. Barrau, J. 1958. Subsistence agriculture in Melanesia. P.Bishop Museum Bull. No.219. Honolulu. Barrau, J. 1963. Plants and the migration of pacific peoples. P.Bishop Museum Bulletin, Honolulu. 117 Barrau, J. 1965. L'humide et sec: an essay on ethnobiological adaptation to contrastive environments in the Indo-Pacific area. Journal of Polynesian Society 74:329-346. Barrau, J. 1965b Witnesses of the past: Notes on some food plants of Oceania. Ethnology 4:282-294. Bayliss-Smith, T. 1977. Human ecology and island populations: The problems of change. In: Subsistence and Survival: Rural Ecology in the Pacific. T.Bayliss-Smith and R. Feachem (eds).pp. 11-23. New York, Academic Press. Beehler, B. 1981. Ecological structuring of forest bird communities in New Guinea. In: Bio geography and Ecology of New Guinea. Vols.l & 2 J.L.Gres.sit (ed.). pp.837-861. London, W.Junk Publishers. Beehler, B. 1985. Conservation of New Guinea rainforest birds. ICBP Technical  Publication No.4 Bellwood, Peter 1978. Man's conquest of the Pacific: The prehistory of Southeast Asia and  Oceania. Auckland, W.Collins Publishers Ltd. Bennett, John 1984 Ecosystem, environmentalism, resource conservation, and anthropological research. In: The ecosystem concept in anthropology. Emilio F.Moran (ed.).pp.289-309. Colorado, Westview Press. Beran, Harry 1980. Massim Tribal Art. Wollongong City Gallery, Australia. Beran, Harry 1988. Betel-chewing equipment of East New Guinea. Aylesbury, U.K., Shire Publications Ltd. Bleeker, P. 1983. Soils of Papua New Guinea. Canberra, Australia, CSIRO and Australian National University Press. Boerboom, J.H.A. and K.F.Wiersum 1983. Human impact on tropical moist forest. In: Man's impact on vegetation. W.Holzner, M.J.A. Werger and I.Ikusima (eds.).pp.83-107 The Hague, Dr.W.Junk Publishers. Brass, L.J. 1959. Results of the 5th Archbold Expedition to New Guinea (1956-57). Bulletin of the American Museum of Natural History 118(1). Brokaw, N.V.L. 1985. Gap phase regeneration in a tropical forest. Ecology 66(3):682-687. Brookfield, H.C. and D.Hart 1971. Melanesia: a geographical interpretation of an island  world. London, Methuen and Co.Ltd. Brookfield, H.C. 1972. Intensification and disintensification. In: Pacific agriculture: a theoretical approach. Pacific Viewpoint 13:30-48. Brush, Stephen 1975. The concept of carrying capacity for systems of shifting cultivation. American Anthropologist 77:798-810. Brush, Stephen 1989. Rethinking crop genetic resource conservation. Conservation Biology 3:19-29. 118 Brush, Stephen 1991. A farmer-based approach to conserving crop germplasm. Economic Botany. 45(2): 153-165. Budelman, Arnoud 1990. Woody legumes as live support systems in yam cultivation. Agroforestry Systems 10:47-59. Bulmer, R.N.H. 1982 Traditional conservation practices in Papua New Guinea. ImTraditional conservation in Papua New Guinea: Implications for today. L.Morauta, J.Pernetta, W.Heaney (eds.). pp.67-76. Boroko, Papua New Guinea, Institute for Social and Economic Research. Chambers, R. 1987. Shortcut methods in information gathering for rural development projects. In: Proceedings of the 1985 International Conference on Rapid Rural Appraisal. Khon Kaen University, Thailand. Chin, S.C. and T.H.Chua 1984. The impact of man on a SE Asain tropical forest. Malaysian  Nature Journal 36:253-269. Chowning, Anne 1959. Witchcraft among the Molima of Fergusson Island. Bulletin of the  Philadelphia Anthropological Society. 12(2). Chowning, Anne 1961. Amok and aggression in the D'Entrecasteaux. American Ethnological Society. Symposium: Patterns of land utilization and other papers. Proceedings of the Annual spring meeting, 1961. pp.78-83. Chowning, Anne 1962. Cognatic Kin Groups Among the Molima of Fergusson Island. Ethnology. 1:92-101. Chowning, Anne 1963. Protomelanesian Plant names. In: Plants and the migration of pacific  peoples. J.Barrau (ed.). Honolulu. P.Bishop Museum Press. Chowning, Anne 1969. Recent acculturation between tribes in Papua New Guinea. Journal  of Pacific History 4:27-41. Chowning, Anne 1977. An Introduction to the Peoples and Cultures of Melanesia. Menlo Park, CA., Cummings Publishing Company. Chowning, Anne 1983. Wealth and exchange among the Molima of Fergusson Island. In The  Kula: New Perspectives on Massim Exchange. Jerry Leach and Edmund Leach (eds.). Cambridge, Cambridge University Press. Chowning, Anne 1989. Death and kinship in Molima. In Death Rituals and Life in the Societies of the Kula Ring. Frederick H. Damon and Roy Wagner (eds.). pp.97-129. Delkab, Northern Illinois University Press. Clarke, W.1971. Place and People: An Ecology of a New Guinean Community. Berkley and Los Angeles, University of California Press. Clarke, W. 1976. Maintenance of agriculture and human habitats within the tropical forest ecosystem Human Ecology. 4(3):247-259. Clarke, W. 1991. Agro-deforestation in Melanesia. Cultural Survival Quarterly. 15(2):45-48. 119 Clawson, David 1985. Harvest security and intraspecific diversity in traditional tropical agriculture. Economic Botany 39:56-67. Clay, Jason. 1988. Indigenous Peoples and Tropical Forests. Massachusets. U.S.A., Cultural Survival Inc. Cohen, Joel, Janis Alcorn, and Christopher S.Potter 1991. Utilization and conservation of genetic resources: international projects for sustainable agriculture. Economic  Botany 45(2V.185-194. Conklin, H. 1957. Hanunoo Agriculture: a report on an integral system of shifting  cultivation in the Philipines. Forestry Dev.Paper 12. Rome, FAO. Conklin, H. 1969. An ethnoecological approach to shifting agriculture. In Environment and  Cultural Behaviour. A.Vayda (ed.). pp. 221-223. New York, Natural History Press. Connel, J. 1978. Diversity in tropical rainforests and coral reefs. Science 199:1302-1310 Conway, Gordon 1987. Rapid rural appraisal and agroecosystem analysis: a case study from Northern Pakistan. In: Proceedings of the 1985 international conference on rapid  rural appraisal.pp. 228-255. Khon Kaen University, Thailand. Dahl, A.L. 1984. Biogeographical aspects of isolation in the Pacific. Ambjo 13(5-6):302-305. Dahl, A.L. 1984b. Oceania's most pressing environmental concerns. Ambio 13(5-6) 296-301. Dasmann, Ray 1983. Biosphere Reserves and Human Needs in Conservation, Science, and  Society. Contributions to the first international biosphere reserve congress, Minsk, USSR. UNESCO-UNEP. Davies,H.L. and D.J. Ivies. 1965. The Geology of Fergusson and Goodenough Islands.  Papua. Geology and Geophysics Report No.82. Canberra, Commonwealth of Australia. De'Ath, Colin 1980. High technology and original peoples: the case of deforestation in Papua New Guinea and Canada. Impact of Science on Society 30(3): 197-209. De'Ath, Colin 1982. Forest conservation practices in Papua New Guinea. In: Traditional  conservation in Papua New Guinea: Implications for today. L.Morauta, J. Pernetta, W.Heaney (eds.). pp.203-215. Boroko, Papua New Guinea, Institute for Applied Social and Economic Research. De'Ath, Colin 1990 Destroying trees, destroying people: forest dwellers' rights in Papua New Guinea. Environments 20(2):64-72. Denslow, J.S. 1980. Gap partitioning among tropical rainforest trees. Biotropica 12(2):47-55. Dewet,J. and Harlan, J.R. 1975. Weeds and domesticates: evolution in the man-made habitat. Economic Botany 29:99-107. 120 Diamond, J.M. 1972. Biogeographic kinetics: Estimation of relaxation times for avifaunas of southwest Pacific islands. Proceedings of the National Academy of Sciences (USA) 69(11):3199-3203. Duduwega, D.K. 1975. Traditions of Nade Village. Oral History 3(4): 1-36. Eaton, Peter 1985 Customary land tenure and conservation in the Papua New Guinea. In: Culture and conservation: the human dimension of environmental planning. J.A.McNeely and David Pitt (eds.). London, Croom Helm. Eaton, Peter 1988. Reinforcing the land ethic: conservation and development through Wildlife Management Areas. In: The ethics of development: choices in development  planning. Papers from the 17th Waigani seminar. Philip J. Hughes and Charmian Thirwall (eds.). pp.65-78. Port Moresby, University of Papua New Guinea Press. Eddowes, Peter 1977. Commercial Timbers of Papua New Guinea. Forest Products Research Centre, Dept. of Primary Industry. Port Moresby. Egloff, Brian 1978. The kula before Malinowski: A changing configuration. Mankind. ll(3):429-35. Egloff, Brian 1979. Recent prehistory in southeast Papua. Terra Australis 4, Department of Prehistory, Research School of Pacific Studies, Australian National University, Canberra. Ellen, Roy. 1975 Non-domesticated resources in Nuaulu ecological relations. Social  Science Information 14:127-150. Ellen, Roy 1978 Nualu settlement and ecology: an approach to the environmental relations  of an Eastern Indonesian community. The Hague, Martinus Nijhoff Publishers. Ellen, Roy 1982. Environment Subsistence and System. London, Cambridge University Press. Fernandez et al 1983. The Chagga homegardens: a multi-story agroforestry cropping system on Mt. Kilimanjaro, Northern Tanzania. Agroforestry Systems 2:73-86. Filer, C.S. and R.T.Jackson 1989. The social and economic impact of a gold mine in Lihir. On file in New Guinea Collection, University of PNG, Port Moresby; Papua New Guinea. Flach,M. and D.L.Schuiling 1989. Revival of an ancient starch crop: a review of the agronomy of the sago palm. Agroforestry Systems 7:259-281. Fortman, Louise 1985. The tree tenure factor in agroforestry with particular reference to Africa. Agroforestry Systems 2:229-251. Fortune, R.F. 1932. The Sorcerers of Dobu: The social anthropology of the Dobu islanders of  the Western Pacific. London, Routledge. Frankel, O.H. 1984. Genetic diversity, ecosystem conservation and evolutionary responsibility. In: Ecology in Practice - part 1: Ecosystem Management F.diCastri et al (eds.). pp.414-427. Paris, UNESCO 121 French, Bruce 1986. Food Plants of Papua New Guinea. Queensland, Australia, Australia Pacific Science Foundation. Gagne, W.C. 1982. Staple crops in subsistence agriculture: their major insect pests, with emphasis on biogeographical and ecological aspects. In: Biogeography and Ecology  of New Guinea Vol.1 J.L.Gressit (ed.). pp.229-258. London, Junk Publishers. Geertz, C. 1970. Agricultural Involution. Berkely, University of California Press. Gideon, Osia. 1990 Personal communication. Senior botanist. Forest Research Institute, Lae, Papua New Guinea. Gleissman, S.R., R.Garcia, and M.Amadora. 1981. The Ecological basis for the application of traditional agricultural technology in the management of tropical agro-ecosystems. Agro-ecosystems 7:173-185. Gleissman, S.R. 1990. Applied ecology and agroecology: Their role in the design of agricultural projects for the humid tropics. In: Race to Save the Tropics: Ecology and  Economics for a Sustainable Future. Robert Goodland (ed.). pp. 33-49. California, Island Press. Gleissman, S.R.(ed.) 1990. Agroecology: researching the ecological basis for sustainable  agriculture. New York, Springer-verlag. Golley, F.B. 1983. Ecodevelopment. In:Tropical Rain Forest Ecosystems: Structure and  function. F.B .Golley (ed.). pp. 335-344. New York, Elsevier Scientific. Gomez-Pompa, A., C.Vazquez-Yanes, and S. Guevara. 1972. The tropical rain forest: a nonrenewable resource. Science 182:1212-1219. GrandstaffT. and S.Grandstaff 1987. A Conceptual basis for methodological development in rapid rural appraisal. In: Proceedings of the 1985 international conference on rapid  rural appraisal, pp.69-89. Khon kaen University, Thailand. Green, E.C. (ed.) 1986. Practicing Development Anthropology Boulder, Westview Press. Gross, Daniel 1984 Ecosystems and methodological problems in ecological anthropology. In: The ecosystem concept in anthropology. Emilio F.Moran (ed.). pp.253-265. Colorado, Westwiew Press. Grynberg, R. and S.Vemuri and D.Wigston 1988. Reconciling ethical behaviour: transfer pricing in the forest industry of Papua New Guinea. In: The ethics of development:  choices in development planning. Papers from the 17th Waigani seminar. Philip J. Hughes and Charmian Thirwall (eds.). pp. 175-204. Port Moresby, University of Papua New Guinea Press. Hames, Raymond 1983. Monoculture, polyculture, and polyvariety in tropical forest swidden cultivation. Human Ecology ll(l):13-33. Harlan, J. 1975. Our vanishing genetic resources. Science 188:618-622. Harlan, J. 1976. Genetic resources in wild relatives of crops. Crop Science 16:329-333. 122 Hart, Robert 1980. A natural system analog approach to the design of a successional crop system for tropical forest environments. Biotropica. 12:73-83. Hayes, T.E. 1983 Ndumba folk biology and general principles of ethnobotanical classification and nomenclature. American Anthropologist Association 85:592-610. Hently,E.E.(ed.) 1981. Handbooks of the flora of Papua New Guinea. Melbourne, Melbourne University Press. Hill, Lance, John Pernetta, and Barney Rongap 1982. The traditional knowledge base: implications for contemporary Papua New Guinea. In: Traditional Conservation in  Papua New Guinea: Implications for today. L. Morauta, J.Pernetta, w.Heaney (eds.). pp.349-363. Boroko, Papua New Guinea, Institute for Applied Social and Economic Research. Hill, Lance, and Philip J. Hughes 1988. Development and the environment. In: The ethics  of development: choices in development planning. Papers from the 17th Waigani seminar. Philip J. Hughes and Charmian Thirwall (eds.). pp. 1-6. Port Moresby, University of Papua New Guinea Press. Holdsworth, David. 1974. Phytochemical survey of medicinal plants of the D'Entrecasteaux. Science in New Guinea 2(2). Holdsworth, David 1975. Medicinal plants from Milne Bay. Science in New Guinea 3(2). Holdsworth, David 1975. Medicinal Plants of Papua New Guinea. Noumea, New Caledonia, South Pacific Commission. Howcroft, Neville 1990. Personal Communication. Section Head, Forest Genetic Resources, Forest Research Institute, Lae, Papua New Guinea. Hyndman, David 1982. Biotope Gradient in a Diversified New Guinea Subsistence System. Human Ecology. 10(2):219-259. Igbozuike, M.U. 1971. Ecological balance in tropical agriculture. The Geographical Review 61:519-529. Ingram, G.B. 1989 Planning district networks of protected habitat for conservation of biological diversity. A manual with applications for marine islands^ yvith primary rainforests. Phd. dissertation University of California, Berkeley. Ingram, G.B. 1990. The need for knowledge from indigenous communities in planning networks of protected habitat for the conservation of biological diversity: three island settings. In Ethnobiology: implications and applications Proceedings of the First  International Congress of Ethnobiology. Vol II. Darell Posey et al (eds.). pp.86-105. Belem, Brazil Ingram, G.B. 1990. Management of biosphere reserves for the conservation and utilization of genetic resources: the social choices. Impact of science on society 158:133-141. Ingram,G.B. and J.T.Williams 1984. In situ conservation of wild relatives of crops. In: Crop  genetic resources: conservation and evolution. J.H.W.Holden and J.T.Williams (eds.). pp. 163-179. London. George Allen and Unwin. 123 Irwin, Geoffrey 1983. Cheiftanship, kula and trade in Massim prehistory. In: The Kula: new  perspectives on Massim exchange. Jerry Leach and Edmund Leach (eds.). pp.29-72 Cambridge, Cambridge University Press. IUCN/UNEP 1986. Review of the Protected Area System in Oceania. Gland, Switzerland, IUCN Jameison, Neil 1987. The paradigmatic significance of rapid rural appraisal. In: Proceedings  of the 1985 international conference on rapid rural appraisal, pp.89-105. Thailand, Khon Kaen University. Janzen, D. 1973. Tropical agro-ecosystems. Science 182:1212-1219. Jenness,D. and Rev.A.Ballantyne 1920. The Northern D'Entrecasteaux. London, Oxford University Press. Johannes, R.E. 1980. Words from the lagoon. Berkeley, University of California Press. Johnson, A. 1973. Individuality and experimentation in traditional agriculture. Human Ecology 1(2):149-160. Kahn, Miriam 1986. Always hungry, never greedy: food and the expression of gender in a  Melanesian society. Cambridge, Cambridge University Press. Kartawinata, K., S. Adisoemarto, S.Risiwan, and A.Vayda 1981. Impact of man on the tropical forest in Indonesia. Ambio 10(2,3): 115-119. Kirch, P.V. 1989. Second Millenium B.C. Arboriculture in Melanesia: Archaeological Evidence from the Mussau Islands Economic Botany 43(2):225-240. Koonlin, Tan 1979. Logging the swamp for food In Sago: the equatorial swamp as a natural  resource. W.P.Stanton and M.Flach (eds.). pp. 13-33. Kuala Lumpur, Malaysia. The Hague, Martinus Nitjhoff Publishers. Kronick, Jay 1984. Temporal Analysis of Agroforestry Systems for Rural Development. Agroforestry Systems 2:165-176. Lakau, Andrew. 1988. Compensation on government land: the Wabag town centre. In: The  ethics of development: choices in development planning. Papers from the 17th Waigani seminar. Philip J. Hughes and Charmian Thirwall (eds.). pp.78-96. Port Moresby, University of Papua New Guinea Press. Lamb, David 1977. Conservation and Management of Tropical Rainforest: A dilemma of development in Papua New Guinea. Environmental Conservation 4(2): 121-129. Lamb, David 1980. Some ecological and social consequences of logging rainforests in Papua New Guinea. Tropical Ecology and Development. 1980:55-63. Lauer, P.K. 1970. Amphlett Islands pottery trade and the kula. Mankind 7(3): 167-176. Lea, David 1972. Indigenous Horticulture in Melanesia. In: Man in the Pacific Islands: Essays on Geographical Change in the Pacific Islands. R.Gerard Ward (ed.). Oxford, Oxford Clarendon Press. 124 Lecroy, M. and W.S.Peckover. A Kalupi 1980. Goldie's Paradise: Display, natural history and traditional relationships of people to the bird. Wilson Bulletin 92(3):289-301. Lecroy, M. W.S.Peckover, A.Kalupi and J.Manseima. 1983. Bird Observations on Normanby  and Fergusson. D'Entrecastreax Island. Papua New Guinea. Wildlife in Papua New Guinea No. 83/1. Boroko, Papua New Guinea, Office of Environment and Conservation. Lithgow,D. and P.Staalsen. 1965. Languages of the D'Entrecastreaux Islands. Port Moresby Department of Information and Extension Services, Territory of Papua and New Guinea. Lovejoy, T.E., J.M.Rankin, R.O.Bierregard, A.B.Rylands, J.R.Malcolm, C.E.Quintela, L.H.Harper, K.S.Brown, A.H.Powell, G.V.N. Powell, H.O.R.Schubart, and M.Hays 1986. Edge and other effects of isolation on Amazon forest fragments. In: Conservation Biology. M.Soule (ed.). pp.257-285. Sunderland, MA, Sinauer Assoc. Luridgren, B. 1982. Introduction. Agroforestry Systems l(l):3-6. Macintyre, M. and M.Young 1983. The persistence of traditional trade and ceremonial exchange in the Massim. In: Melanesia: beyond diversity. Vol. 1 R.J.May and Hank Nelson (eds.). pp.207-233 Canberra, Research School of Pacific Studies, ANU. Malinowski, B. 1922. Argonauts of the Western Pacific. London, Routledge and Kegan. Malinowski, B. 1935 Coral Gardens and their Magic. Vol.1 London, Allen and Unwin. Manner, I. 1981. Ecological Succession in New and Old Swidden Gardens of Montane Papua New Guinea Human Ecology 9(3):359-377. Mckinnon, John andKathy, G.Child, J.Thorsell. 1986. Managing protected areas in the  Tropics. IUCN Meggers, B. 1971 Amazonia: Man and culture in a counterfeit paradise. Chicago, Aldine. Mergen, Francois 1987. Research opportunities to improve the production of homegardens. Agroforestry Systems 5:57-67. Michon,G. J.Bompard, P.Hecketsweiller and C.Ducatillion 1983: Tropical forest architectural analysis as applied to agroforests in the humid tropics: the example of traditional village-agroforests in West Java. Agroforestry Systems. 1:117-129. Michon,G., R.Mary and J.Bompard 1986. Multistoried agroforestry garden system in West Sumatra, Indonesia. Agroforestry Systems 4:315-338. Miracle, M.P. 1967. Agriculture in the Congo basin. Madison. University of Wisconsin Press. Moran, Emilio 1984. Limitations and advances in ecosystems research. In: The ecosystem  concept in anthropology Emilio F. Moran (ed.). Colorado, Westview Press. Nair, P.K 1985. Classification of agroforestry systems. Agroforestry Systems 5:97-128. 125 Nelson,H. 1976. Black. White and Gold: Goldmining in Papua New Guinea 1878-1930. Canberra, Australian National University Press. Newton, Douglas. 1975 Massim. Museum of Primitive Art, New York. Ninez, V. 1984. Household gardens: theoretical considerations on an old survival strategy. Food Systems Research Series, No.l. Lima, International Potato Center. OHier,C.D. and C.F.Pain 1980. Actively rising surficial gneiss domes in Papua New Guinea. Journal of the Geological Society of Australia 27:33-44. Oilier, C D . and C.F.Pain 1981. Active gneiss domes in Papua New Guinea: New tectonic landforms. Zeitschrift fur Geomorphologie 25(2): 133-145. Oilier, C D . and C.F.Pain 1988. Morphotectonics of Papua New Guinea Zeitschrift fur  Geomorphologie. Supplement 69:1-16. Padoch, Christine and Wil de Jong 1991. The house gardens of Santa Rosa: diversity and variability in an Amazonian agricultural system. Economic Botany 45(2): 166-175. Paijmans,K. 1975. Vegetation Map of Papua New Guinea Melbourne, Australia. CSIRO. Paijmans, K. 1976. Vegetation. In: New Guinea Vegetation. K.Paijmans (ed.). Canberra, Australian National University Press. Paijmans, K. 1979. Ecological notes on sago in New Guinea. In: Sago: the equatorial swamp  as a natural resource. W.P.Stanton and M.Flach (eds.). Proceedings of the 2nd International Sago Symposium. Kuala Lumpur, Malaysia. Pain,C. F.Kensington, and C D . Oilier. 1981. Geomorphology of a Pliocene granite in Papua New Guinea. Zeitschrift fur Geomorphologie 25(3):249-258. Panoff, F. 1972. Maenge gardens: a study of Maenge relationships to domesticates. Phd dissertation. Canberra, Australian National University. Papua New Guinea Bureau of Statistics. 1981. National Census 1980: Milne Bay Province Preliminary Field Counts Port Moresby, Papua New Guinea Bureau of Statistics. Peet, R.K. and D.C Glenn-Lewin, J.Walker Wolf 1983. Prediction of man's impact on plant species diversity: a challenge for vegetation science In: Man's impact on vegetation W.Holzner, M.J.A.Werger, and I.Ikusima (eds.). London, Dr.W.Junk Publishers. Percival, Margaret and J.S. Womersley. 1975. Floristics and Ecology of the Mangrove  Vegetation of Papua New Guinea. Botany Bull.#8, Lae, PNG, Dept. of Forests, Division of Botany. PernettaJ. and Hill,L. 1982. International pressures on resource management in Papua New Guinea. In: Traditional Conservation in Papua New Guinea: Implications for Today. L.Morauta, J.Pernetta, W.Heaney (eds.). pp. 319-332. Boroko, Papua New Guinea, Institute for Applied Social and Economic Research. Pigram,C. and H.L. Davies. Terranes and the accretion history of the New Guinea orogen. Journal of Australian Geology and Geophysics 10:193-211. 126 Posey, D. 1990 The application of ethnobiology in the conservation of dwindling natural resources: lost knowledge or options for the survival of the planet. In: Ethnobotany:  implications and applications. Proceedings of the First International Congress of  Ethnobiology. Vol I. Darren Posey et al (eds.). pp.46-59. Belem, Brazil Posey, D. 1985. Indigenous management of tropical forest ecosystems: the case of the Kayapo indians of the Brazilian Amazon Agroforestry Systems. 3: 139-158. Powell, J.M. 1976. Ethnobotany. In: New Guinea Vegetation K.Paijmans (ed.). pp.106-183. Canberra, Australian National University Press. Powell, J.M. 1977. Plants, man, and environment in the Island of New Guinea In: The  Melanesian Environment. J.H.Winslow (ed.). pp.11-21. Canberra, Australia, Australian National University Press. Powell, J.M. 1982 History of man's plant use and man's impact on the vegetation. In: Biogeography and Ecology of New Guinea Vol.1 J.L.Gressit (ed.). pp.207-227 The Hague, Junk Publishers. Power, A. 1988. Resource development in East Sepik province. In: The ethics of development: choices in development planning. Papers from the 17th Waigani seminar. Philip J. Hughes and Charmian Thirwall (eds.). pp.270-281. Port Moresby, University of Papua New Guinea Press. Prance, G.T. 1984. Introduction on knowldege and use of the Amazon forest. Interciencia 9(5):275-279 Pratt, T.K. 1981. Biogeography of birds of New Guinea. In: Biogeography and Ecology of  New Guinea. Vols.l & 2 J.L.Gressit (ed.).pp.815-836. London, W.Junk. Prescott-Allen, R. and C.Prescott-Allen 1982. The case for in situ conservation of crop  genetic resources. Nature and Resources 18:15-20. Purseglove, J.W. 1972. Tropical crops Vol 1 & 2 London, Longman Group Ltd. Raintree, J.B. 1986. Agroforestry Pathways Unasylva 38(154). Rappaport, R. 1967. Ritual Regulation of Environmental Relations Among a New Guinea People Ethnology 6:17-30. Rappaport, R. 1974. The flow of energy in an agricultural society. In: Biological Anthropology (Readings from Scientific American) S. Katz (ed.).pp.371-387. San Fransisco, W.H.Freeman. Reddel,P. and P.A.Robrook and P.A.Ryan. 1989. Managing Nitrogen fixatoin in Casuarina species to increase productivity. In: Trees for the tropics: growing multipurpose trees  and shrubs in developing countries. D.J.Boland (ed.). Canberra, Australian Centre for International Agricultural Research. Rhoades, R. 1985. Breaking new ground: Agricultural anthropology. In: Practicing Development Anthropology Edward Green (ed.).pp.22-67. Boulder, Westview Press. 127 Rhoades, R. 1987. Basic field techniques for rapid rural appraisal. In: Proceedings of the  1985 international conference on rapid rural appraisal, pp. 114-129. Thailand, Khon Kaen University. Sabhrasi,S. 1978. Effects of forest fallow cultivation on forest production and soil. In: Farmers in the forest: economic development and marginal agriculture in northern  Thailand. P.Kunstadter, E.C.Chapman, and S.Sabhasri (eds.).pp.l60-184. Honolulu, The University of Hawaii Press. Sack, Peter. 1975. Mythology and land rights on Wogeo. Oceania 56(l):40-52. Seddon, G. 1984. Logging in the Gogol Valley, Papua New Guinea. Ambio 13(5-6):345-50. Seligman, Carl G. 1910. The Melanesians of British New Guinea. Cambridge, Cambridge University Press. Siltoe, Paul. 1983. Roots of the Earth: Crops in the highlands of Papua New Guinea. Manchester, Manchester University Press. Skar, H.O. (ed.) 1985. Anthropological contributions to planned change and development. Goteborg, Sweden, Acta Universitas Gothoburgensis. ' Smith, E.A. 1979. Human Adaptation and energetic efficiency. Human Ecology 7(l):53-74. Soemarwoto, Otto 1987. Homegardens: a traditional agroforestry system with a promising future. Agroforestry: A decade of development. Howard A.Steppler (ed.) 158-170 Nairobe, ICRAF. Tabora, P.C. 1991. Analysis and evaluation of agroforestry as an alternative environmental design in the Phillipines. Agroforestry Systems 14:39-63. Thaman, R.R. 1988. Fijian agroforestry: trees, people, and sustainable polycultural development. In: Rural Fiji J. Overton (ed.) Suva, Fiji, Institute of the Pacific Studies, University of the South Pacific. The Government of Papua New Guinea. 1975. The Constitution of the Independent State of  Papua New Guinea. Government Printer, Port Moresby. Thelen, K.D. and Child,G.S. 1984. Biosphere reserves and rural development: In: Conservation. Science and Society 470-478 UNESCO-UNEP. Tolhurst, N.M. 1951. The origin of man and woman: a native legend from Fergusson Island. South Pacific 5:132-133. UNEP 1976. Ecodevelopment. Report of the Governing Council. Fourth Session. Nairobi (UNEP/GC/80). Unruh, Jon D. 1990. Iterative increase of economic tree species in managed swidden-fallows of the Amazon. Agroforestry Systems. 11:175-197. Vayda, A.P. and R.Rappaport 1968. Ecology, Cultural and Non-cultural. In Introduction to  cultural anthropology. J.Clifton (ed.) Boston, Houghton and Miffling. 128 Vayda, A.P. 1983. Progressive contextualization: methods for research in human ecology. Human Ecology 11(3):265-281. Vasey, D.E. 1985. Household gardens and their niche in Port Moresby. Papua New Guinea. Food and Nutrition Bulletin 7:37-43. Vergara, N. 1985. Agroforestry Systems: A Primer UnaSylva 37(147). Vergara, N. and P.K.R.Nair. 1985. Agroforestry in the South Pacific region - an overview. Agroforestry Systems 3(4):364-379. Vickers, William 1983. Tropical forest mimicry in swiddens: a reassessment of Geertz's model with Amazonian data. Human Ecology ll(l):35-45. Viner, A.B. 1984. Environmental protection in Papua New Guinea. Ambio 13(5-6):342-344 Waddell, E. 1972. The mound builders. Seattle, University of Washington Press. Wagner, Roy 1989. The exchange context of Kula In: Death Rituals and Life in the Societies  of the Kula ring. Frederick H.Damon and Roy Wagner (eds.).pp.255-280 Delkab, Northern Illinois Press. Waiko, John 1977. The people of Papua New Guinea, their forests and their aspirations. In The Melanesian Environment. J.H.Winslow (ed.) 407-429 Canberra, Australian National University. Waiko, John and K.Jiregari 1982. Conservation in Papua New Guinea: custom and tradition. In Traditional Conservation in Papua New Guinea: implications for today L.Morauta, J.Peretta, W.Heaney (eds.). pp.21-38. Boroko, Papua New Guinea, Institute for Applied Social and Economic Research. Webb, L.J. 1982. The human face in forest management. In Socio-economic Effects and  Constraints in Tropical Forest Management E.G.Hallsworth (ed.). pp. 159-175. New York, John Wiley and Sons Ltd. Weiner, Annette B. 1976. Women of value, men of renown: new perspectives in Trobriand  exchange. Austin, University of Texas Press. .. Weinstock, J.A. 1984. Tenure and Forest Lands in the Pacific. Working paper-SEast-West Center, Honolulu, Hawaii. Weinstock, J.A. and Vergara, N.T. 1987. Land or plants: agricultural tenure in agroforestry systems. Economic Botany 41:313-322. White, K.J. 1977. Constraints on developing forest industries in Papua New Guinea. In: The  Melanesian Environment J.H.Winslow (ed.).pp.407-429. Whitmore, T.C. 1969. Geography of the flowering plants. In A discussion on the results of  the Royal Society Expedition to the British Solomon Islands Protectorate, 1965. Phil. Trans. R. Soc. Land., B,255,549 Whitmore, T.C. 1973. Plate tectonics and some aspects of Pacific plant geography. New  Phvtol pp.1185-1190 129 Whitmore, T.C. 1984. Tropical forests of the far east. Oxford, Oxford Clarendon Press. Whisder, Arthur 1988. Tokelauan Ethnobotany. Economic Botany 42:155-175. Wiersum,K. 1982. Tree Gardening and Taungya on Java: examples of agroforestry techniques in the humid tropics. Agroforestry Systems 1:53-70 Williams, C.N., W.Y. Chew, J.H.Rajaratnam 1979. Tree and field crops of the wetter  regions of the tropics. London, Longman Group Ltd. Wood, A.W. 1981. The soils of New Guinea. In: Biogeography and ecology of New Guinea. Vol. 1 and 2 J.L. Gressit (ed.).pp.73-83. London, W.Junk Publishers. Wood, A.W. and G.S.Humphreys 1982. Traditional soil conservation in Papua New Guinea. In: Traditional Conservation in Papua New Guinea: Implications for Today. L.Morauta, J.Pernetta, W.Heaney (eds.).pp.319-332. Boroko, Papua New Guinea, Institute for Applied Social and Economic Research. Womersley, J.S.(ed.) 1978. Handbooks of the Flora of Papua New Guinea Vol.1 Melbourne, Melbourne University Press. Yen, D.E. 1973. The origins of Oceanic agriculture. Archeology and Physical Anthropology  in Oceania. 8:68-85. Yen, D.E. 1982. The history of cultivated plants. In: Melanesia Beyond Diversity Vol 1. R.J.May and Hank Nelson (eds.) Canberra, Australia, Australian National University Press. Young, M.W. 1971. Fighting with Food: Leadership, values and Social control in a Massim  society. London, Cambridge University Press. Young, M.W. 1987. Wapolu Gold Mining Project: a socio-economic impact study. ANUTECH Pty Ltd. and Dept. of Anthropology RSPacS, Australian National University, Canberra. 130 APPENDIX A Notes on Nade plant classification There may be some inconsistencies when trying to relate Nade classification to the Latin binomial system. Some botanically identified plants may have more than one local name. Some local names may be names of a species at a different growth stage. Some local names may group a few botanical species under one name. Repeated questioning eliminated discrepancies as much as possible. Those plants in Appendix 2 that are given botanical names are verified to exist at Nade. Confusion in classification also arises from the local use of both Nade and Dobu language. Informants often didn't know whether a plant name they had given was a Dobu name or Nade name. Some local plant names have a meaning others don't. Some may have had a meaning in Nade language but that meaning has been forgotten. Identifying features may be included in the name. For instance, some trees are recognized by habitat such as mountain mangrove or cuscus'oya. In Nade, plants may be grouped by use, by form, or by smell. A general name for everything eaten as greens is Pal Gobuwala refers to several broadleaf climbers. Gau refers to the smell of Cinamomum spp. and is also the name of the tree. Often the name of a category becomes a nickname for the plant, used when the other name is forgotten. The name is doubled if it is a wild plant that is not used and there is a similar domesicated or semi-domesticated plant that is used. For example Saido nut and sctidosaido, or tewa and tewatewa. Names of medicinal and magic plants may be given as nicknames such as "our dusty grandmother" or "our grandmother's necklace". These are names old people use if they are sending a child to collect them and they don't want to use the real name. These plants should also have a real Nade name. However, the secrecy surrounding those plants made it difficult to learn the corresponding name. While the people of Nade identify several kinds of plants to species level (for instance Pandanusl and to variety for yams, sweet potatoes, and taro, some plants they broke down into different names for the same species. 131 APPENDIX B Table Bl Ethnobotanical inventory Name commonly used in Nade. Other local names or common name. Botanical Name Successional Phase and life form. Use Aigwama Acalvpha hispida Brum Euphorbiaceae * xni introduced shrub. Medicinal - leaves boiled taken internally for dysentry. Ornamental - leaves boiled with leaves of Pandanus spp. for black dve for woven mats Aikwado-kwadoia kwadokwadoia II tree Food - vine leaves cooked for greens. Anelewa Cananga odorata Hook. Annonaceae * VII, HI, VI . tree . Medicinal - fiber from twigs tied around the neck. Ornamental - perfume made from flowers and clarified coconut oil, and applied during dancing. Apo'apoa Costus speciosa (Koen) J.K.Sm. Zingiberaceae * in, II shrub Medicinal - taken internally to heal womb after birthing. Apose'ulana gavagava (trobriands) Scaevola taccuda Goodeniaceae ** XII shrub Medicinal - leaves crushed into salt water and taken internally as a contraceptive. Leaves boiled and taken internally to kill worms. 132 Table Bl cont' Ataito'ala Selaginella velutina Casati Selaginaceae * II, V, VII, VI, III fern ally Medicinal - Mothers boil leaves and drink as tea when breastfeeding for babies health. Magic - leaves worn in hair as protection from forest spirits. Babakamitawa II shrub Medicinal - leaves boiled and taken internally for fever and worms. General tonic for babies. Badala VI tree Construction - axe handles (hardwood) Baubau dumo (dobu) Bambusa sp. Construction - flooring, fencing. Bawelalade Myristicaceae * VI, VII tree Construction Bawelo'u II vine Medicinal - leaves taken internally after delivery to stop bleeding. Ornamentation Bedinegwa II, vm shrub Medicinal - leaves combined with bunala, tabunomeso, and apuapua and washed with to reduce fever. Beluwai Sphaerostephanos alatellus (Christ.) Copel. Thelypteridaceae * n, vii fern Food - fronds cooked and eaten as greens. 133 Table B l Conf Bidu bidu Barringtonia papuana Laut. Lecytidaceae * i, ii, xni,in, v i small tree Living fences, land-marks, considered sacred when planted at abandonned housesites. "When it flowers it tells them that the wild fowl will lay eggs. They recognize another edible species by the same name Bolia-momohawa bush yams II vine Food - tubers eaten during poor yam harvest. Bolobolo wild taro Hunting - fed to dogs to make them wild, aggressive, and sensitive for hunting wild pig. Bolo'usi - Smilax leucophvlla Bl. Liliaceae * II vine Medicinal - leaves chewed with betelnut as a contraceptive; leaves boiled and taken internally as a general tonic. Bo'unadi mominawa (Dobu) II vine Food - tubers eaten during famine. Budowai budo Alstonia scholaris Apocynaceae **** n, v, vi, vn tree Construction - posts, beams Commercial species Bwasibwasi Sphaerostephanos invisus (Forst.F.)Holtt. Thelypteridaceae * n, VII, III fern Medicine - leaves rubbed on babies stomach for stomaches. Magic. 134 Table 1 cont' Cuscus oya mountain mangrove VI tree Construction. Food - edible fruits. Dabadaba Scleria polvcarpa Boeck. Cyperaceae * I, II sedge Medicinal - abortificant Dabedabe wea (dobu) Alpinia sp. Zingiberaceae * n large herb Medicinal - young leaves wrapped inside leaves of Dagwena'u heated over fire then applied externally to infected sores. Dabedabe dim dim xin shrub Introduced from highlands. Believed to frighten snakes away from housesites. Daewa breadfruit Arctocarpus altilis Moraceae *** in, v, vii, n tree Food - fruits eaten raw. Medicinal - young shoots chewed for concentration while carving, or painting canoes. Dagia Cvathea angiensis Cyatheaceae *** Vm (600-2200m) tree fern Construction - good posts. Dagulala galauto vii, ii, vm tree Construction. Commercial tree species Dagwena'u in, II shrub Seasonal indicator. Food - leaves used for mumu cooking (a system of cooking in which yams are wrapped in leaves and surrounded by hot rocks). 135 Table 1 cont' Damaia kalakai(dobu) Ficus copiosa Steud. Moraceae * n, V I I tree Food - fruits eaten cooked. Medicinal. Deba'o Crataevareligiosa Forst.f. Capparidaceae * X V shrub Gardening - used for yam seed magic. Ornamental - flowers for decoration. Dedepia Alpinia spp. Zingiberaceae * n, in, xm herb Ornamental - leaves crushed into coconut oil to make perfume for dancing. Dekwakwa ka'elabikwa (dobu) Saurauia defaurii c.f.(F.Muell) Diels Saurauiaceae * I I tree Medicinal - mothers boil leaves and drink as tea after birthing to prevent blood clots and clean the womb. Dibi Laportea interupta L. Chev. Urticaceae ** I I , I I I , V shrub Medicinal - rubbed on skin to relieve muscle fatique. Diwelowelo n vine Ornamental - black hair dye. Dubedube I I vine Medicinal - roots rubbed on gums to relieve toothache. Magic - attracts women; also prevents yams from escaping from garden. Edonene Neonauclea chalmensii (F.V.M.)Merr. Rubiaceae * X I V shrub Medicinal - leaves boiled or crushed into water and taken internally as general tonic May also be used as contraceptive. 136 Table B l cont' Egauda II vine Construction - string for tying fences and posts. Egegela Decaisninia hollrungii (K.Sch.) Leigh Loranthaceae * II epiphyte Parasitic epiphyte. Said to kill yam plants. Ekwelikweli II tree Construction. Elima VII, II herb Magic - shoots chewed for immunity to love spells. Food - leaves used for wrapping and cooking yams. Enoenososo Dennstaedtia penncilifera v.A.v.R. Dennstaedtiaceae *-n, in fern Food - seasoning for clay pot cooking. Medicinal. Enwalalade n fern Magic - for enhancing voice. Gabowa Schefflera stahliana (Warb.) Frolin Araliaceae * VI small tree Bird of paradise roost. Magic - Leaves used for yam washing yam and for covering yams during feasts to increase their distribution. Saplings used as hunting traps. 137 Table Bl cont' Gaga morabau Crinum asiaticum L. Amaryllidacea ** xin, xn shrub Medicinal - roots and young shoots chewed as general tonic and disease prevention. Fruits roasted in coconut leaves and rubbed in hair to get rid of lice. Fishing - inner stem used as bait. Gagawala pilopilo (dobu) Ficus septica Burm f. Moraceae * II, Ul, VII tree Medicinal - leaves heated on fire and rubbed on body for fever, or applied to boils and swellings, or rubbed on stomach to reduce enlarged spleen from malaria. Cooking - neutralizes toxicity of poisonous fish, such as red emporer. Moraceae tree Gala and heated over fire to condense the sago. Ganona Amvdrium spp. ** Araceae Medicinal - coughs, cuts and swellings Gau Cinnamomum sp. Lauraceae VII, VI tree Important in sorcery. (personal communication, Osia Gideon, senior botanist, Forest Research Institute, Lae). Gewalagilili Amvema scandens (Tiegh.) Dansen Loranfhaceae * II tree Medicinal - leaves boiled and taken internally for coughs. 1 3 8 Table B l cont' Giligili Planchonella (trade name) Planchonella ateridifera H.J.L. (White and Francis) Sapotaceae * n, V I I tree Commercial timber species. Construction - good for housebeams. Seasonal indicator - when fruit falls then it is time for wildfowl to lay eggs. Gita'ana'alaia II vine Medicinal - leaves washed with to cure pneumonia. Yam magic Gobuwala II, VII, VI vine Medicinal - leaves applied externally to new cuts; heated on fire and applied to swellings. Gogamagasddi keletenadi (dobu) walabi'ie II vine Medicinal - leaves crushed into water and taken internally to prevent worms, or when bones are stuck in the throat. Young shoots roasted and eaten for dairhea. Gogom Planchonia papuana Knuth. Barringtoniaceae * VII, V tree Construction - softwood, good for houseposts. Magic - Inner bark used for washing yam seeds. Goma'ala Commersonia bartramia L.Merr Sterculiaceae * n tree Construction - stems used for houseposts. Individuals claim exclusive rights. Goma'uwe II vine Hunting - fed to hunting dogs improve sense of smell. 139 Table B l conf Gulamweda Ficus pungens Reinwex Bl. Moraceae * II tree Medicinal - inner bark boiled and taken internally for pneumonia. Top shoots squeezed into water and drunk to relieve stomach pain. Gwagwa'ula XV shrub Food - cooked with fish to neutralize toxins. Gwautobulo'a II, VII tree Medicinal - young shoots boiled and fed to babies "to make young them fat". Gwasi Pipterus argenteus (Forst.)Wedd. Urticaceae * n shrub Utensil - leaves used for cleaning cooking pots Gwede'u II, VII tree Resin for repairing canoes. Himo sebulu Pandanus sp. II tree Construction - aerial roots used as roof beams. Iabwa VII tree Construction. Ike chestnut Inocarpus edulis Leguminaceae *** xii, ni, xni tree Food - nuts. Iku'iku II, VII shrub Medicinal - leaves boiled and taken internally for coughs or tuberculosis. 140 Table B l cont' Ilimo Terminalia sp. **** Combretaceae VI tree Construction - hardwood used for carving knife and axe handles. Yam magic. Pig medicine Ipuluma II fern Ornamental - black dye for grass skirts. Itaita n herb Medicinal - leaves taken as tea as general tonic. Kaibwadubwadue II, VII tree Food - edible fruits. Kaigalisaladi n tree Construction - flooring. Kaigobu II shrub Food - seasoning for clay pot cooking. Kakolu Hvdriastele kasesa Palmaceae ** v, m palm tree Construction - stem used for flooring. Kaladimumu Ficus wassa Roxb. Moraceae * II, III, V small tree Magic- tonic for babies. Leaves worn as protection from forest spirits. Kalaneania Ficus sp. II tree Food - leaves cooked and eaten. 141 Table B l con? Kalo kalo noko obstusifolia R.Br. Verbenaceae ** Premna tree n, m drunk for pneumonis Medicinal - leaves crushed into water and drunk Kamagili'gilile n, VII vine Medicinal - for stomach ache and dairrhea. Kamwakuku citrus III, XIII tree Food - edible fruits. Kapukapula Rejoua aurantiaca Gaud. Apocynaceae * II, III tree Construction. Kapwei II. tree Food - edible fruits. Kasakasa Ficus sp. Moraceae II tree Medicinal - leaves taken internally for coughs, sore throats and dysentry. Kasawe mango Mangifera spp. Anacardiaceae *** ni, xiii tree Food - fruits eaten raw. Kasawe tomatai wild mango n,VII tree Food - fruit. Kasikasi'oya II tree. Construction - for house posts. 142 Table Bl cont' Kasiyala black palm V tree Construction - flooring, roofing, axehandles, outriggers, spear shafts Magic. Food - inside of shoots eaten. Ownership applied. Katakata'iuna Ficus insculpta summeh. Moraceae * n tree Magic - lips painted with crushed leaves to negotiating power in Kula trading. Ornamental - Mix leaves with uyoyo or sinodolia flower, paint on skin and it peels off in a temporary tatoo. Kawagigi VI tree Construction - good for house posts. Kawaino Cvathea contaminaus Cyatheaceae *** vm tree fern Construction - posts. Bark fibers used to make pillows. Kebulua II tree Garden digging stick. Kebulibuli n tree Carving - axehandles. Kedewa'eu'eudi Cvathula prostrata (L)Br. Amaranthaceae * XIII herb Medicinal - flowers and leaves boiled and taken internally for dysentry. 143 Table B l cont' Kedikedidi Melastoma affine D.Don Melastomaceae * I I shrub Garden digging stick. Kedikwai mulumulu (Dobu) Kleinhovia hospita L. Sterculiaceae ** I I small tree Utensil - fire flint, digging stick. Medicinal - crushed leaves applied externally to treat skin fungal disease. Kediodiole Leucosvke capitellata Poir. Urticaceae * I I shrub Utensil - used as a pot scrubber Kegalu tulip Gnetum gnemon L. Gnetaceae * ii, I I I , vn, xm tree Food - leaves cooked and eaten. Construction - bark fiber for tying fences and house materials. Medicinal. Kekawa Gvmnacrantera paniculata (D.C.)Warb. var. Zippeliana (Miq.)Sincl. Myristaceae * I I , vn tree Hunting - fiber used as string for tying spears. Kekula gai I I , V I tree Fishing - used to make nets. Kekwasala I I I , xm shrub Ornamental - perfume to anoint the body during dancing. 144 Table Bl cont' Kemwata Euodia hortensis J.R., G.Forst Rutaceae * m, xm shrub Ornamental - leaves crushed with coconut oil to make perfume to anoint the body during dancing. Sacred - planted at graveyards. Keninitu m, xm shrub Ornamental - Perfume. Sacred - Planted at graveyards. Kepo'u malakwa (dobu) Leguminaceae I I , I I I tree Clothing - Cambium used to make grass skirts. Keyakwayakwala m herb Medicinal - chew young leaves to feel light. Kipukipua Ficus botrvocarpa Miq. Moraceae * I I tree Construction - soft wood, but good outhouse withstands salt water Medicinal - fruit juice squeezed onto open sores. Kokauta I I vine Medicinal - leaves taken internally for sinus colds. Kokone pesa pesa Pennisetum macrostachvum (Brongn) Trin. Gramineae * i, n grass Medicinal - leaves boiled and taken internally, shoots to relieve sore throat. Kokowita n, V I I tree Medicinal - inner shoots chewed as general tonic. Food - edible fruits. 145 Table Bl conf Kopuleta Horsfeldia septica (Roxb)Sincl. var.septica Myristicaceae * II tree Food - fruits eaten raw, young shoots eaten cooked. Kubakuba'sesela uba'ubasesela alanuma (dobu) Oreocnide trinervis (Bl.) Miq. Urticaceae * II, V, III tree Medicinal - crushed leaves taken internally for headache or feeling hot. Construction - fence posts. Kukupi II small tree Medicinal - leaves boiled and taken internally for indigestion. Kulaikulaia II shrub Medicinal - tea for losing weight Magic - Placed under yams at feast so yams will be distributed further. Kulisa II, VII tree Food - edible fruits. Kwakwamowala kwakwan'oya VI, VIII tree Construction. Kwakwamu Calophvllum inophvllum L. Guttiferae ** XII tree Medicinal -Magic - used for travelling at night. Construction - posts and beams. Kwanakwana-matadi Brevnia racemosa Muell.Arg. Euphorbiaceae * II, HI, XIII shrub Ornamental - boil to make black dye. Tobacco Magic 146 Table Bl cont' Kwasio Leguminaceae II, VII tree Fishing and hunting - sap used to poison spear tips. Kwa'usa Glochidion novogmnense K.Sch. Euphorbiaceae * II tree Construction - canoes carved from stem; resin used to repair leaks. Labia Sago Metroxvlon sagu Palmaceae *** V tree Construction - thatching of roofs and walls. Food - starch from stem. Medicine - stem sap applied to the forehead to ease headaches. Laku VII tree Food - edible fruits. Lamakekiti II, in herb Medicinal - leaves taken internally for pneumonia. Lamalama Canavalia rosea (S.W.)D.C. Leguminaceae * XII herb Medicinal - young leaves of female flower are crushed into water and taken during second stage of birthing to ease labour. Also for treating tuberculosis. Lamoita Sida rhombifolia L. Malvaceae * II herb Medicinal - roots chewed or brewed as tea for Legilegi VII tree Construction. 147 Table Bl cont' Lekunoleku vn tree Construction. Leuceana no local name Leuceana sp. Leguminaceae III tree Introduced for shade crop for cocoao in the 1970's Lite budagaia n, vn vine Construction - fiber used for tying fences or housing. Loanna Yagwana Albizzia falcatoria (L.) Fosb. Leguminaceae * II, VI tree Construction. Medicinal - inner bark fed to pigs to eliminate worms. Lobolobo in, herb perfume Lopepwaisa kala pwaipwai'ala (dobu) Hopea sp.**** VI big tree Construction - house posts and beams. Lucoluco Heliconia bihai (L.) Musaceae ** II shrub Food - leaves for wrapping yams for mumu cooking. Lulu VI, VII tree Construction. Commercial species. Lulululu sago thorny Metroxvlon rumphii Palmaceae *** V tree Construction - thatching for roofs and walls. Ornamental - young sago leaves shredded and dried to make grass skirts for dancing. Utensil - ribs of fronds make broom handles. Table Bl cont' 148 Mabo II, VII tree Pest control - Use sticky sap to spread on branches or fruits of tree to trap birds. Maimaibola Ageratum conzoides L. Compositae * II herb Medicinal - rub young leaves on cuts, on sores, or on lymph glands. Maluwana Red cedar Toona sp. Meliaceae **** II tree Construction - house posts and beams; canoes Matabau II tree Medicinal - leaves crushed and sniffed for blocked sinuses. Mataupo VI tree Construction - hardwood good for paddles and posts. Ornamental - yellow dye. Mawgwamawgwa Bobowaiuwa Athvrium silvaticum (Bl.)Milde Athyriaceae * II fern Food - seasoning for clay pot cooking. Meguwa Sanapu (Normahby) Sida cordifolia Malvaceae ** II shrub Fishing - seeds crushed into stream to poison freshwater fish. Mekelegalepa idiwaige II vine Construction - stem fiber used for tying houses and fences. Me'u kwila Intsia sp. V, VI tree Construction - house posts and beams, carving canoes. Milemilele gigowai Pandanus sp. III, V tree Ornamental - used to decorate grass skirts. 149 Table B l cont' Mogaiya n, v n vine Construction - string to tie houses and fences Mwadaidai gigaku Graptophvllum pictea (L.) Griff Acanthaceae * xm shrub Medicinal. Mwadawa rosewood (trade name) Pterocarpus indicus Leguminaceae * n, in, vi , v n tree Construction - posts, canoes Commercial species Medicinal - wash with leaves to cool the body. Mwagolu mwakiko Svgium mallaccense (L.)Merr. Perry Myrtaceae * n, III tree Food - fruits eaten raw. Medicinal - women wash with crushed leaves to make them strong for giving birth. To speed recovery they wash with a mixture of mwagolu, yagayaga, and ubaubasesela. Mwahadi wild ginger Zingiber sp. Zingiberaceae ** ii, ni, vn, XIII herb Magic - source of all magic spells. Adds power to spells. Protects against evil forest spirits. Three types recognized: white for healing; red for killing; mwahadimwahadi for eating. Mwenana II, VII tree Construction Naiwa nalea (dobu) notea Macaranga densifolia Warb. Euphorbiaceae * II tree Construction - used for house posts Medicinal - leaves crushed and rubbed on open sores. Pregnant women wash it. Nanagala Pandanus sp. V tree Household - leaves used for weaving floor mats for houses. 150 Table B l cont' Nauma'owaga wagawaga Crassocephalum crepidoides (Benth.)S.Moore Compo sitae ** II, I shrub Medicinal - leaves crushed and applied to sores. Nawa II shrub Ornamental - grass skirts made from stem fibers. Nimabadi lala (dobu) Ervthrina variegata (L.) Leguminaceae * xm, xn tree Medicinal - antidote for stonefish poisons: bark used to rub on skin. Nipunipuna Lvgodium tripercalium ** II vine Ornamental - stems used as string to weave armlets. Construction - string to tie fences. Nwanuwano Poikilonospenaen amboinensis Zipp. Urticaceae * n, VII climber Eaten by bird of paradise. Medicinal - leaves squeezed into water and taken internally for coughs and asthma, headaches or overheating. Paikwado'kwadoia Polvporandra scandens Becc. Icacinaceae * XV, vn, II tree Food - leaves boiled and eaten as greens Paiuwa dalele (nickname meaning-bitter) Pangium edule Reinw. Flacourtiaceae * II tree Food - fruits cooked and eaten Panawasi II shrub Food - greens. 151 Table B l cont' Pesaisa Pteris pacifica Hieron. Pteridaceae * I I fern no use Piesi (Dobu) Cordvline fruticosa LA.Chev Liliaceae * ii, I I I , xm shrub Cooking - laid over yams in clay pot for seasoning. Pilili pedidi (Normanby) Hvptis suaveolens Pott. Labiatae ** ii. vni herb Medicinal - leaves applied to cuts and wounds. Ornamental - perfume Pilipilili Calaphanoides aff. angusta (WartY) Brem. Acanthaceae I I , I I I herb Potea potea baia baiawa Steganthera salomonensis (Hensl.) Phillips Monimiaceae * I I , X V tree Young leaves chewed as stimulant like betelnut. Pupuwelo X I I vine Magic - to make people invisible. Pwasio I I shrub Magic - forest spirit protection. Pwatapwata X I grass Medicinal - Rubbed on throat when a fishbone is stuck. 3 152 Table Bl cont' Sagusagu beach pandanus Pandanus sp. XII shrub Functions as moth balls when woven into grass skirts. Prevents rats and bugs from eating skirts. Salewa Pandanus sp. XII tree Food - edible fruits. Saido curry berry okari nut Terminalia kaernbachii Combretaceae* * * * * xm, in tree Food - nuts eaten raw or roasted. Saidosaido red saido II, VI tree Construction -Sakasakayaina n vine Medicinal - Leaves rubbed on forehead for headaches. Roots chewed for sterility. Sakwikwi kwikwi Ficus hispidoides S.Moore Moraceae ! II ! tree Medicinal - leaves crushed on new cuts to function as a coagulant. Sebosebo Pandanus sp. II tree Construction - Used as string to tie sago thatching and flooring and spears. Selala n shrub Medicinal - wash with leaves for energy. Selaseld ubusudi n shrub Perfume. 153 Table B l conf" Senala Pandanus sp. II tree Fishing - line made from aerial roots. Sila II shrub steelwool. Siliputa Desmodium heterocarpon (L.)DC. Leguminaceae* n . herb Medicinal - roots chewed as permanent contraceptive. Sinate Pandanus sp. VII, V Hair comb made from roots. Sinuwa dillenia (commercial trade name) Dillenia castaneifolia Dilleniaceae * VI tree Construction. Commercial species Medicinal - mothers wash with leaves after giving birth to heal the womb". Sinudilia Mussaenida procera Bail. Rubiaceae * II, I shrub Magic - flowers used to paint lips orange to increase negotiating power during Kula trading. Sitabiti Pandanus sp. xm, III, XII shrub Ornamental - woven into mats introduced from Fiji Sitasita VI shrub Medicinal - leaves used to wash with as a general tonic. Soba gomakalameyadi Asplenium cvmbofolium christ. Aspleniaceae * n, VI, VII HI, V epiphyte Medicinal - mothers drink as tea after birthing Food - young leaves boiled and eaten as greens. 154 Table Bl cont' Sopu Dendrobium bifalce Lindl. Orchidaceae i ii, vn : epiphyte Ornamental - stem fibers woven into armlets. Sukwasukwa II tree Perfume. Suyala red type Timonius timon Rubiaceae ** VII shrub Medicinal - leaves good for pneumonia Magic - used in black magic spells. Suyala white type Timonius sp. II, VII shrub Construction. Tabunameso means "never dies" n vine Medicinal - leaves crushed and rubbed on sores. Magic - for yams. And to neutalize spells. Taginamanamala Gvmnacranthera paniculata (D.C.)Warb. Myristicaceae * vn, v i tree Utensils - used to carve canoe paddles. Saplings used to make spear shafts. Tali tali Rubus mollucanas L Rosaceae * I, II shrub Medicinal - chewed and spit on stiff joints. Tamwenau VII, II tree Construction -Taptapisala II fern Utensil - Washing cooking pots. 155 Table B l conf Tawana Taun (commercial trade name) Aphanamixis polvstachva (Wall)Park Meliaceae * VI tree Construction - used for house posts Tawatawa Wedelia biflora Compositeae ** II herb Medicinal - to treat enlarged spleens from malaria and also for intestinal worms. Also protection from forest spirits. Tawana-wialele II, VII tree Construction Tewatewa Piper sp. Medicinal - for worms or malaria symptoms. To'ana'esinua to'uand bagula'etaula Litsea globosa Kost. Lauraceae * II tree Gardening - leaves used as green manure for sugarcane. Tobatobala bunala (Dobu) (Roxb.) Amomum aculeatum Zingiberaceae * II, ni and headaches. Sap shrub Medicinal - used to treat coughs, sores, wounds taken internally as as antidote to poisons. Internal stomach cleanser. Cooking - Cook with paiuwa fruit to neutralize fish poisons. Tobolomina II, VII tree Construction. Totoba Macaranga quadriglandulosa Warb. Euphorbiaceae * VI tree Construction - stem good for house posts: root fibers used as posts: root fibers used as filler for repairing canoes. Fruits eaten by Goldie's bird of paradise. 156 Table Bl cont' Totosi ii, vii, ni tree Food - edible fruits Tua Derris uliginosa (Benth.) (Fortune 1932) II, III, VI vine Medicinal - roots swabbed on open sores; acts as anti-bacterial. Fishing - roots crushed and placed in streams to poison fish. Tubunaikauna II, VII vine Medicinal - roots chewed with betelnut as a contraceptive. Tumwesi sumwesi II, VI, VII tree Construction Tupwotupwo II tree Construction Tutuwana ulabo Alstonia spectabilis ** v, v n tree Construction - hardwood for canoes. Medicinal. Ubuga Callipteris prolifera (Lam. )B org Athyriaceae * II, vn fern Food - boiled and eaten as greens. Ulo'ulo II, VII tree Construction - hardwood for carving axehandles. 157 Table Bl cont' Utaba Medusanthera laxifolia Icacinaceae * VI tree Food - mature leaves used for mumu cooking; young leaves cooked and eaten. Fruit eaten by bird of paradise (P.decora) (Lecroy et al 1980). Construction. U'upi II, VII tree Medicinal - rub on breasts to ease breast feeding. Used to stop daihhrea in infants. Uyoyo Nephorolepis hirsutala (Forst)Presl Nephoroleaceae* II, xm fern Medicinal. Magic. Waigaugau bwesobweso Lauraceae **** Cinnamomum sp. tree vn,v roots chewed to treat Medicinal - leaves boiled and taken internally or tuberculosis. Construction- carving canoes. Wakawaka VI tree Hunting - chew on roots to improve eyesight for hunting possums at night. Waku magimagi 11, VI palm Construction - flooring, roofing Wakuwaku n, V, VII tree Food - young shoots are eaten. Waligobogobo Spafhoglottis plicata XI, II herb no use Bl. Orchidaceae * 158 Table B l conf Wa'otu (Nade) piesi (Dobu) Cordvline fruticosa L.A.Chev Liliaceae * ii, HI, xni shrub Cooking - laid over yams in clay pot for seasoning. Wa'otu (Nade) piesi (Dobu) Cordvline terminalis Kunth. Liliaceae I I , I I I , X I I I Landmarker. Wasewasegala V , V I I shrub Food - seasoning. Wasi wasi Lvgodium trifurcatum Bak. Schizeaceae * I I , I H , V I vine Ornamental - stem fiber is used to weave armlets for decoration during dancing. Wasemu Cassia alata L. Leguminaceae ** Medicinal - leaves are squeezed until soft and rubbed on body to treat sipoma, a fungal disease. We'a'apimua I I shrub Medicinal - Mothers drink as tea after one month of pregnancy as a health tonic. Wenasi I I tree Construction. Wetui Flagellaria indica L. Flagellariaceae* I I , V I I , V I vine Medicinal- Leaves boiled and taken internally for pneumonia; young shoots rubbed on gums for toothache, or taken internally for dairrhea. Wibwaga V I tree Construction. 159 Table B l cont' Wilai adibai tewa Piper sp. ** n, m vine Medicinal - leaves boiled and taken as tea for constipation; roots chewed as contraceptive. Yagayaga II tree Construction - timber Medicinal - leaves washed with after delivery to prevent blood clots and to stop bleeding. Yasiyasi II shrub Perfume Yewawa nanamaia (Dobu) nanabeyama Homalanthus novoguinensis (Warb.)Laut.&Sch. Euphorbiaceae * n, in tree Medicinal - leaves crushed into water and washed with as protection from pneumonia; or dropped into clear sinuses. Drink as tea to clear womb after delivery. Yowai Hibiscus tiliaceus L. Malvaceae * II tree Construction - inner bark used as fiber for tying houses. Medicinal - leaves made into drink to help in birthing, or painful menstration wild taro Feed to dogs to make them wild and aggressive, and sensitive for hunting. Gardenia cf. gjellerupii Val. Rubiaceae * Food of Goldie's Bird of Paradise. * From plant collection on file at PNG National herbarium, Forest Research Institute. **Holdsworth(1974) *** French (1986) **** Eddowes (1977) ***** Archibold Expeditions (1953) 160 Table B2 Vernacular list of botanically unidentified plants of undetermined use. Name most Successional Phase Vegetation Profile # used in Nade and plant form Baiabaiawa H,VI tree 20 Bawakwawadi VI seedling 28 Budobudowala II tree 8 Bwasianasikai tree 29 Dalele II tree 7 Ibwanono VI vine 12 Dimomo II shrub 13 E'ape'ana II herb 21 Ega'uda II vine 16 Egulegule II tree 17 Ekwaulo VII seedling 33 Eple'e II seedling 7 Esinesinea'ala n , v n 33 Gwautobuloa II tree 17 Iki Utree 5 Ipuyeta VII tree 18 Iyayosa Pandanus sp. Kagopa II shrub 27,24 Kaibwasobwasoge VII tree 29 Kedowasaladi II vine 16 Kekelipwaia Utree 21 Kekowago VII herb 29 Ke'ologe VI tree 28 Kukumyuyudi vine 18 Kulabwaio II tree 8 Lelewai VI tree 28 Matabwe'ala VI tree 28 Matasapasapaila VII tree 29 Molumolu VI herb 28 161 Table B2 conf Mwagolmwagol II, VII tree Mwanamwana II vine Mwomwone VI tree Opi'e II fern Osiyo'u II tree Sema II Pandanus spp. Soiso II tree Suabu'oya VI tree Tawanaladi II tree Tenagasigasi VI tree Tubuda'ana'mwagula II palm yanamatan II tree Yowasi VII fern Wabugala II vine Waduwadukai II vine 162 29 5 28 2 13 8 16 12 27 28 16 27 29 16 17 APPENDIX C  VEGETATION COMMUNITY DATA Table CI DATA FOR 1990 SWIDDENS Gardener Topo graphy Area (ha.) Elevation (m) Age (years) Average stump size (cm.) Soil type * Slope % #adults I.Eugene slope 0.09 120 10-15 14 A 40-60% 1 2.Eugene coastal plain 0.23 10 6 10 D 0 1 3. Lucie ridge crest 0.04 160 6-9 13 A 10% 1 4. Patrick upper slope 0.08 160 6-9 13 A 10% 1 5.Ema coastal plain 0.15 10 6 10 D 0 1 6.Mele coastal plain 0.18 10 6 11 D 0 1 7. Imanuel lower slope 0.33 80 10 18 A 30% 2 8. Jack ridge 0.12 260 18 20 A 30% 1 9.Pascal ridge 0.08 250 18 20 A 35% 1 163 Gardener Topo graphy Area (ha.) Elevation (m) Age (years) Average stump size (cm.) Soil type * Slope % #adults 10. Ekilon upper slope 0.62 320 20 up 25 A 40% 2 11. Raphael Sonyan Jane midslope 0.39 110 10 11 A 30 3 12. Janes mother upper slope 0.16 175 12 13 A 35% 1 13.Ben Enosoa lower slope 0.09 50 15 10 D boulders 10% 1 14. James lower slope 0.36 40 8-10 7 D boulders 15% 2 15.Benon midslope 0.08 50 13 . 14 A 1 16.Caleb midslope 0.13 230 11 12 A 30% 1 17. Virginia ridge 0.11 320 15 18 A 25-30% 1 18. Famela lowerslope creekside 0.21 60 15 15 D 10% 1 164 Gardener Topo graphy Area (ha.) Elevation (m) Age (years) Average stump size Soil type * Slope % #adults 19. Kaola midslope 0.38 260 40 up 50 B 25% 2 20. Diwisi coastal plain 0.32 20m 20 up 36cm D flat 2 21.Rhoda upper slope 0.34 340 20 A no rock 40-60% 2 22.Peter uncle midslope 0.09 70 5 6 A no rock 25% 1 23 .Peter midslope 0.12 110 8 9 A no rock 30% 1 24.Tony Goroti lower slope 0.14 50 12 9 A rocky 10% 1 25. Aiseya coastal plain 0.8 20 8 10 C sandy flat 2 26. Raymond coastal plain 0.31 20 10 15 C sandy flat 1 * A Eutropepts, dystropepts and troporthents on igneous and metamorphic rocks. B. Vitrandepts of volcanic origin. C. Tropofluvents. D. Fluvaquents. Average garden size per adult: 0.17 ha. Range of garden size: 0.04 to 0.8 ha. Average length of fallow: 12.8 years Range of fallow length: 5-40 years 165 Table C2 Summary of vegetation profiles Profile # Successional Phase and age (years) Topography and elevation Site factors slope (%) and soil type * Total # plant species # plant species named # plant species used Informants sex age 1. fallow (5) lower slope ridge 120m 25% A 20 13 13 F 12 2. fallow (4) lower slope ridge 110m 5% A 25 20 12 F 31 3. fallow (10) lower slope 160m 30% • A .19 16 11 M 26 4. fallow (9) midslope , 230m . 40% A 25 16 16 F 28 5. fallow (6) ridge 160m 50% A 25 21 20 F 28 6. hamlet agroforest coastal plain flat C 30 30 28 F 50? 7. hamlet agroforest coastal plain flat alluvial C 34 34 29 F 35 8. fallow (12-15) coastal plain flat alluvial B 25 24 18 M 20 166 Profile # Successional Phase and age (years) Topography and elevation Site factors slope (%) and soil type * Total # plant species # plant species named # plant species used Informants sex age 9. mixed sago groves coastal plain flat alluvial C 22 21 15 M 20 10. fallow (6) footslope 130m flat B 25 20 18 M 40 11. modified hill forest upper slope 370m 45% A 35 25 11 M 28 12. fallow (15) coastal plain 50m flat alluvial C 33 32 24 M 40 13. fallow (10?) mid-slope 220m 45% A 30 23 17 M 28 14. fallow (12?) lower slope 150m 35% A 31 22 12 M 28 15. fallow (9) coastal plain flat C 31 23 19 M 28 16. fallow (10) coastal plain flat C 39 34 27 M 35 17. fallow (12) footslope rolling 60m 20% B 24 21 14 F 30 167 Profile # Successional Phase and age (years) Topography and elevation Site factors slope (%) and soil type * Total # plant species # plant species named # plant species used Informants sex age 18. fallow (30) footslope 30m 5% alluvial C 31 31 25 F 30 19. mixed sago groves coastal plain 10m flat alluvial C 16 16 15 M 25 20. fallow (30) footslopes 350m 15% B 26 26 18 M 60 21. fallow (30) midslope 250m 10% B . 33 32 21 M 60 22. fallow (10) coastal plain 60m , flat C 32 32 21 23. fallow (22) ridgecrest 160m 30% A 26 22 18 M 18 24. fallow (13) ridgecrest 200m 30% A 31 23 18 M 18 25 hamlet agroforest footslope 50m 20% A 35 33 28 M 18 26. fallow(7) ridgecrest 150m 25% A 25 25 18 F 25 168 Profile* Successional Phase and age (years) Topography and elevation Site factors slope (%) and soil type * Total # plant species # plant species named # plant species used Informants sex age 27. modified hill forest ridgecrest 550m 35% A 33 10 5 M 35 28. modified hill forest upperslope 500m 35% A 39 20 15 M 30 29. lowland forest coastal plain 30m flat, alluvial C 30 26 25 F 50 30. mixed sago groves coastal plain 10m flat, alluvial C 20 20 18 F 30 31. Gravesite ridgeside 330m 30% A 36 17 10 F 25' 32. hamlet agroforest coastal plain 20m 10% C 32 32 27 M 35 33. lowland forest coastal plain 20m 5% C 36 30 22 M 35 34. lowland forest footslope 80m 15% C 34 28 24 F 30 * A Eutropepts, dystropepts and troporthents on igneous and metamorphic rocks. B. Vitrandepts on volcanic footslopes. C. Tropofluvents. D. Fluvaquents 169 APPENDIX C VEGETATION PROFILES VILLAGE AGROFOREST (Fig.5) Profile 6 Informant(s): Sinatawai History: gardened 1980 Location: 200m NE of mission station Elevation: 10m Topography: Coastal plain Soils: Alluvial Canopy closure: 50% Dominants: Arctocarpus sp., Mangifera sp., Ficus spp., Pterocarpus indicus. Structure: Layered; dense herb and shrub layer with emergent trees, 2-3 tree layers. Components: Use A. Budowai (Alstonia scholar!?) B. Dabedabe (Alpinia sjx) C. Uyoyo fern (Nephorolepsis hirstula) D. Naiwa (Macaranga densifolia) E. Kegalu (Gnetum gnemon) F. Kwa'usa leguminaceae G. Paiuwa (Pangium edule) H. Kwanakwanamatadi (Breynia racemosa) I. Sila Urticaceae J. Gai K. Bidubidu (Barringtonia papuana") L. Wasi wasi M. Mwadawa (Pterocarpus indicus) N. totoba (Macaranga quadriglandulosa) O. Kasawe (Mangifera minor) P. Daewa (Arctocarpus altilis) Q. Sakwikwi (Ficus hispidoides) R. kalaneania (Ficus sp.) S. Tewa (Piper sp.) T. Gwabagwaba (Ficus sp.) U. Dagwenau V. Kalokalo (Premna obtusifolia) W. Dibi (Laportea interupta) X. Wetui (Flagelleria indica) Y. Gwasi (Pipteru argeneus) Z. Ataitoala (Selaginella velutina) a. Beluwai (Sphaerostephanos alatellus) b. coconut (Cocos nucifera) d. kipukipua (Ficus botryocarpa) e. Bulo'usi (Smilax leucophylla) f. kopuleta (Horsfeldia septica) g. mwahadi (Zingiber sp.) construction medicinal medicinal food construction food ornamental technology technology land-marker construction construction food food medicinal food food food seasoning medicinal medicinal medicinal technology magic edible fern food construction medicinal fruits magic 170 Fig. 5 Schematic profile of hamlet agroforest. 171 MIXED SAGO GROVE (Fig.6) Profile 30 Liformant(s): Eugene History: managed for sago production since ? Location: 200m east of mission station, 100m from beach Elevation: 10m Topography: Coastal plain Soils: Alluvial Canopy closure: 65% Dominants: Arctocarpus sp., Planchonella aeteridifera, Pterocarpus indicus Components use A. Budowai (Alstonia scholaris) construction B. kakolu construction C. Sago (Metroxvlon sagu) construction D. Potepotea (Pangium edule) fruit E. Mwadawa (Pterocarpus indicus) construction F. Bwasibwasi (Sphaerostephanos invisus) G. Gaga (Crinum asiaticum) medicinal H. Giligili (Planchonella ateridifera) construction I. Soba (Asplenium cvmbifolium) medicinal J. Gobuwala K. Nipunapuna (Lvgodium tripercalium) ornamentation L. Gogom (Planchonia papuana) construction M. Wasewasegalu fibre N. himo (Pandanus sp.) fibre O. Daewa (Arctocarpus sp.) fruit P. Damaia (Ficus copiosa) fruit O. Coconut (Cocos nucifera) fruit R. Tutuwana (Alstonia spectabilis) construction S. Gagawala (Ficus septica) medicinal 172 30<". Fig.6 Schematic profile of mixed sago grove. i!2> FALLOW MOSAIC (Fig. 7) Profile 17 Informant(s): Imanuel History: gardened 1978 Location: E. of Bwaubwau creek Elevation: 75m Topography: Rolling coastal plain Soils: Alluvial, volcanic origin Canopy closure: 40% Dominants: at 18m, Gala Ficus sp., Kepo'u leguminaceae Components: Use A. Kepo'u Leguminaceae grass skirts B. kasiyala black palm construction C. Kipu kipua (Ficus botrvocarpa) construction D. Gwautobuloa medicinal E. Tewa (Piper sp.) food F. Tewa tewa (Piper sp.) G. Tumwesi H. Kukupi medicinal I. Mwenana construction J. Bwasibwasi (Sphaerostephanos invisus) medicinal K. Egulegule medicinal L. Gagawala (Ficus septica) M. Wetui (Flagellaria indica) construction N. Diwelowelo ornamental O. Kebulibuli technology P. Mwadawa (Pterocarpus indicus) construction Q. Asamabuta R.Bodaboda S. Matabau medicinal T. Sakwikwi (Ficus hispidoides) medicinal U. Kebunadodo V. Daewa (Arctocarpus altilis) food W. Waduwadukai X. Posiu Y. Yagayaga construction, medicinal Z. loanna construction a. gwabagwaba (Ficus sp.) medicinal b. Twanawiawialele construction d. Bolibolia food e. Dagwenau (Alpinia sp.) ornamental f. Waku g. Kwa'usa leguminaceae construction h. kebukebuku i. kalaneania (Ficus sp.) food j. gualauto k. shrub 1. kediodiole (Leucosvke capitellata) digging stick m. weluya n. ekwelikweli 174 p. pntp.potea fPangium edule) food q. tamwanau r. gobuwala medicinal s. pote potea t. kekawe fishing nets 175 Fig.7 Schematic profile of fallow phase. 176 MODIFIED LOWLAND FOREST (Fig. 8) Profile 29 Informant(s): Mele Location: Between Mania an Waipe creeks Elevation: 30m Topography: coastal plain Soil: Alluvial, very rocky, CL History: disturbed, never cleared Canopy closure: 80% Dominants: Pterocarpus indicus, Myrsticaceae, Ficus spp., Alstonia scholaris, Tumwesi. Components: Use A. Tumwesi construction B. gobuwala medicinal C. bawelalade (Myristicaceae) construction D. dibila construction E. matasapasapaila F. bolobolo G. broadleaf fern H. mwadeuwa - small tree greens I. unidentified seedling J. totosi K. iawana (Ficus sp.) food L. kakolu construction M. bwasianasikale N. paidwadodwadoia (Polvporandra scandens) food O. mwagol mwagol (Szvgium sp.) construction P. giligili (Planchonella aeteridifera) Q. Gonawa R. dagulala construction S. seedling T. sapling U. fern V. kaibwasobwasoge - black dye ornamental W. kekowago X. tewatewa (Piper sp.) Y. dabedabe (Alpinia sp.) medicinal Z. kasiyala black palm construction a. potepotea (Pangium edule) food b. yowasi 177 MODIFIED SMALL-CROWNED HILL FOREST (Fig.9) Profile 28 Informant(s): Pascal Location: Edagwaba Elevation: 500m Topography: 35% slope, dissected hillside Soil: CL History: may have been gardened more than 50 years ago Canopy closure: 75% Dominants: up to 40m. Dillenia spp., Albizzia falcatoria, Pterocarpus indicus Structure: herb layer dominated by ferns and tree seedlings, heavy vine layer, layered Components: Use A. wibwaga construction B. bawakwawadi C. cuscus oya D. seedling E. seedling construction F. Sinuwa (Dillenia castaneifolia) G. Potepotea (Pangium edule) food H. bawelalade (Myristicaceae) construction I. seedling J. seedling K. seedling L. kedibulele - (Ficus sp.) M. seedling N. suyala oya O. shrub P. shrub Q. shrub R. bush fern S. molumolu medicinal T. sapling U. tenagasigasi V. sapling construction W. sebosebo palm X. taginamanamala (Gvmnacranthera paniculata) construction Y. seedling Z. seedling a. seedling b. fern d. fern e. gabowa (Schefflera stahliana) Bird of paradise food f. wetali g. kwakwamu oya - big tree h. kegalu (Gnetom gnemon) greens i. mwomwone k. budobudo m. seedling n. ke'ologe q. metabwe'ala p. seedling 179 Fig.9 Schematic profile of modified .small-crowned hill forest. APPENDIX D Map 7 Map of Inventory survey location APPENDIX E. Questionnaire for an informal guided interview I. Plant Inventory 1. What is this plant called? 2. Does that name have a meaning? 3. Does this plant have another name? 4. What is this plant used for? 5. What part of the plant is used? 6. How is it used? 7. Does it have another use? 6. Where is it usually found? 7. Do you collect parts of it or all of it? 8. What parts? 9. Do you ever plant it? 10. Do you plant seeds? 11. Do you plant a part of the plant? 12. Does someone own it? 13. Can anyone harvest from it? 14. Who will inherit it? 15. When do you harvest it? II. Gardening Techniques 1. Why do you burn your garden? 2. Why do you lay logs? 3. How do you choose where to plant the Kwatea siakutu taro cassava banana pineapple aibika sugar cane papaya 4. Do you think this plant is growing well? why or why not? 5. Would it grow better somewhere else? III. Land tenure 1. Who cleared this garden? 2. Where are the boundaries? 3. How are the boundaries marked? 4. Who planted the markers? 5. How did you know about the markers? 6. How did you choose this site? 7. Could anyone else garden on this site if you didn't? 8. Whose yam seeds are/will be planted here? 9. Who will take the harvest? 10. Will you garden on this site again? If so when? . 11. Who has the rights to the site after the harvest? 12. Can other people enter the site? 13. Who gave you rights to this garden? 14. Does this garden have a name? 15. Who named it that? I V . Land-use decisions 1. How did you choose where to make your garden? 2. Did you discuss it with anyone else? Who? 3. Did you decide on your own? 4. When did you choose this site? 5. How does this site compare to your last year garden? 6. How does this years crop compare to last years garden? 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items