LICUATI FOREST RESERVE, MOZAMBIQUE: FLORA, UTILIZATION AND CONSERVATION

Transcription

1 LICUATI FOREST RESERVE, MOZAMBIQUE: FLORA, UTILIZATION AND CONSERVATION by Samira Aly Izidine Research Project Report submitted in partial fulfilment of the requirements for the taught degree Magister Scientiae (Systematics and Conservation Evaluation) in the Faculty of Natural & Agricultural Sciences Department of Botany (with Department of Zoology & Entomology) University of Pretoria Pretoria Supervisor: Prof. Dr. A.E.van Wyk May 2003

2 I LICUATI FOREST RESERVE, MOZAMBIQUE: FLORA, UTILIZATION AND CONSERVATION Samira Aly lzidine 2003

3 To the glory of God and to the memory of my dear father, Aly Abdul Azize Izidine, A gl6ria de Deus e em memoria ao meu querido pai, Aly Abdul Azize Izidine,

4 TABLE OF CONTENTS DEDICATION... i LIST OF FIGURES AND TABLES... iv LIST OF FIGURES AND TABLES... v ABSTRACT... 1 PROJECT PROPOSAL... 3 Rational and motivation... 3 Principal aims... 5 Research questions... 6 Layout of Report... 6 CHAPTER 1. GENERAL INTRODUCTION Background Physical description Climate Flora and vegetation of Maputaland Vegetation Flora and endemism Forests ofmaputaland Sand Forest Forest reserves in Mozambique Licuati Forest Reserve History and location Flora and vegetation SABONET expedition New record of a Xylopia species CHAPTER 2. HISTORICAL BACKGROUND General The Santaca family Interview with Chief Santaca Medicinal Plants The Sacred Forest Utilisation Economic and social development CHAPTER 3. GENERAL MATERIAL AND METHODS Visit to the study area Field work

5 3.3 Herbarium studies Checklist of plants of Licuati Thicket Dynamics of Licuati Thicket Historical information and current utilisation oflicuati Thicket Studies onxylopia (leaf anatomy) Leafmacromorphology (SEM studies) CHAPTER 4. THE LICUATI THICKET Title page Abstract Introduction Material and methods Results and discussion Conclusions Recol111llendations Acknowledgements References CHAPTER 5. IDENTITY OF XYLOPIA TORREIN. ROBSON Introduction Material and methods Conclusions Acknowledgements Specimens examined References CHAPTER 6. SUMMARY GENERAL ACKNOWLEDGEMENTS CURRICULUM VITAE GENERAL REFERENCES APPENDICES APPENDIX 1: ANNOTATED CHECKLIST OF PLANTS OF LICUATI THICKET APPENDIX 2: MISCELLANEOUS PHOTOGRAPHS Ill

6 LIST OF FIGURES AND TABLES Figures Figure 1. Map showing the location of the Licuati Forest Reserve (LFR), southern Mozambique Figure 2. Map showing the Licuati Thicket, Licuati Forest Reserve, southern Mozambique Figures 3-6. SEM micrographs of leaf lower surfaces. 3 & 4 Xylopia torrei, 5 & 6 Xylopia odoratissima Figures SEM micrographs of leaf lower surfaces. 7 & 8 Xylopia parviflora, 9 & 10 Xylopia tomentosa Tables Table 1. Number of families, genera and species ofpteridophyta, Monocoty ledonae and Dicoty ledonae recorded from the Licuati Thicket... : Table 2. Synopsis of families that contribute 1% or more to the total number of species in the Licuati Thicket, LFR Table 3. Synopsis of families that contribute 1% of more of the total number of MC endemics in the Licuati Thicket, as compared to the figures for the Maputo Elephant Reserve Table 4. Families with five or more small tree or shrub species in the flora of Licuati Table 5. Comparison between Xylopia torrei, Xylopia parviflora and Xylopia odoratissima IV

7 LIST OF ABBREVIATIONS DNPH = Dinitropheny lhydrazine EtOH = Ethanol FAA = Formalin-acetic-acid-alcohol GIS = Geographical Information System GMA = Glycol methacrylate HEMA = 2-Hydroxyethyl methacrylate IUCN = World Conservation Union LFR = Licuati Forest Reserve LMA = National Herbarium, Maputo LMU = Eduardo Mondlane University Herbarium, Maputo LISC = Lisbon Herbarium, Lisbon MC = Maputaland Centre of Endemism PRE = National Herbarium, Pretoria PRU = H.G.W.J. Schweikerdt Herbarium, University ofpretoria SABONET = Southern African Botanical Diversity Network WWF = World Wide Fund for Nature V

8 ABSTRACT Licmiti Forest Reserve, Mozambique: flora, utilisation and conservation by Samira Aly Izidine Research Project Report Submitted in partial fulfilment of the requirements for the taught degree Magister Scientiae (Systematics and Conservation Evaluation) Supervisor: Prof. Dr. A.E. van Wyk A floristic study of the Thicket vegetation in the Licmiti Forest Reserve, southern Mozambique, was undertaken. Observations on the structure and range of this unique vegetation type, its species composition, level of endemism and utilisation are provided. A literature review of the physical environment, soils, climate, vegetation and flora, as well as historical background, is presented. The Licuati Thicket covers an area of about ha, which is about 35% of the ha covered by the Licmiti Forest Reserve [LFR]. A study of satellite images and aerial photographs, taken over a period of about 30 years between 1958 and 1989, indicate that no significant changes took place in the size and distribution of the Licmiti Thicket during this period. Attributes such as common names, life history, growth form, fruit type and medicinal uses are provided for 113 plant taxa. These are presented as a checklist arranged according to family, with genera and species listed alphabetically within the families. The Licmiti Thicket contains about 13% of the vascular plants endemic to the Maputaland Centre of Endemism [MC]. The plant families Rubiaceae, Fabaceae and Celasteraceae contribute the largest numbers of endemics.

9 The first record of Xylopia torrei N.Robson from southern Mozambique is reported. An amplified description is provided for the species, supplemented by notes on the comparative anatomy of the genus in southern Africa. The Licmiti Forest contains about 45 plants that are utilized by humans for medicinal purposes. The plant parts most commonly collected are leaves, roots and bark. The traditional leadership of the Santaca family still plays an important role in the conservation of the Licmiti ecosystem. An interview was conducted with members of the Santaca family, and some of the historical and cultural information obtained from this interview are presented in this report. This study provides information that supports the recognition of the Licmiti Thicket as a unique vegetation type distinct from Licmlti (Sand) Forest. Both these vegetation types are endemic to the MC. I trust that this report will raise public awareness of the uniqueness of this special botanical region in Mozambique and that it will guide reserve management and sustainable utilisation programmes in the LFR. 2

10 PROJECT PROPOSAL (condensed) 1.1 Rationale and motivation In 1994, southern Mozambique and the northern part ofkwazulu-natal were recognised by the WWF/IUCN as a centre of plant endemism, called the Maputaland Centre [MC] (Davis et al ). High levels of endemism, spread across virtually the whole taxonomic spectrum, including both plant and animal taxa, were highlighted (Van Wyk & Smith 2001 ). The MC covers the coastal plain of southern Mozambique and northern KwaZulu-Natal, South Africa. Several conservation areas have been proclaimed in the area, including the Greater St Lucia Wetland Park (a World Heritage Site), Tembe Elephant Park, Maputo Elephant Reserve, Licmiti Forest Reserve, Mkuzi Game Reserve, and Kosi Bay Coastal Forest Reserve (Van Wyk & Smith 2001). The vegetation of the MC is very diverse, and includes different types of grassland, bush veld, thicket, forest and swamp vegetation. Two remarkable vegetation types, Sand Forest and Woody Grassland, are endemic to the MC (Van Wyk & Smith 2001). It is estimated that at least vascular plant species occur in the MC (Van Wyk 1996; Van Wyk & Smith 2001), and that at least 230 species or infraspecific taxa are endemic/nearendemic to the region. Three genera, Brachychloa, Galpinia and Helichrysopsis are endemic/near-endemic to the MC. The plant families Asclepiadaceae, Euphorbiaceae, Rubiaceae, Liliaceae, Acanthaceae and Asteraceae contain the largest numbers of endemic species. The Licmiti Forest Reserve [LFR] in southern Mozambique, was established on 14 December 1943 by portaria number 5354 (Gomes e Sousa 1968). It is located between the Tembe and Maputa rivers, and the Porto Henrique-Bela Vista and Bela Vista-Catuane roads, at 26 20' and 32 30'. The reserve and surrounding areas cover about ha (Mangue 1999). 3

11 A rare thicket vegetation type, locally referred to as the Licuati, occur in the LFR. The Licuati Thicket contains many rare plant and animal species. It is very distinctive, not only because of its unique combination of plant and animal species, but also because it is more-or-less confined to the ancient coastal dunes of northern KwaZulu-Natal and the extreme southern parts of Mozambique. Because of its restricted occurrence and unusual species compliment, Licuati Thicket is one of the most important plant communities in the MC (Van Wyk & Smith 2001 ). In the past, Licuati Thicket has been classified together with Licuati Forest as a single vegetation type sometimes also called Sand Forest. The Licuati ecosystem plays an intricate part in the lives and livelihoods of local communities; providing land for agriculture, timber, wild foodstuffs, building material, medicine, fuel wood, fodder, and household commodities and services. It is also a source of cultural symbols and sacred sites (Mangue 1999). Although the LFR was created in 1943 to protect woody plant species, particularly Afzelia quanzensis, the enforcement of laws on timber and charcoal production is very weak, resulting in the destruction of some of the vegetation types in the reserve. Overexploitation of traditional medicinal plants is largely the result of commercialisation in urban areas, a breakdown of cultural taboos, and the introduction of more efficient harvesting implements such as cane knives and metal axes. Deforestation around the Forest Reserve due to charcoal production threatens not only the loss of plant resources, but also the indigenous knowledge systems of several endemic species (Mangue 1999). The LFR lies within the development zone of Maputo, and is under increasing pressure from extensive harvesting of wood for the production of charcoal. Since the area around the forest has been resettled, the structure and diversity of large tree species have declined notably (Bandeira, unpublished results). The population of Maputo is estimated to be about (INE 1997). Studies by the National Directorate of Forest and Wildlife suggest that over 80% of the Maputo population depend on fuel wood for household cooking and heating purposes. It is estimated that the annual consumption of firewood and charcoal by the urban population of Maputo City is 4

12 about 12 million tons/year. This demand is placing considerable pressure on forest resources particularly around the LFR. Presently there is no enforcement of conservation measures at LFR. Hypothesis 1. Licuati Thicket is a distinct vegetation type, quite different from Licuati Forest. Hypothesis 2. Licuati Thicket is rich in plant diversity. Hypothesis 3. A significant proportion of species in the Licuati Thicket are MC endemics/near endemics. Hypothesis 4. The species of Xylopia recorded from Licuati Thicket is conspecific with X torrei. Hypothesis 5. The structure of the Licuati Thicket has changed due to timber and charcoal production. 1.2 Principle aims To record the historic information about the Santaca family, Licuati Forest and its utilisation by local population. To comply a checklist of vascular plants oflicuati Thicket. To annotate the checklist by adding various attributes (biological and ecological), including common names, growth form (tree/shurb/climber, herb), life history (perennial, evergreen or deciduous), fruit type (fleshy or dry) and medicinal uses to the taxa included in the checklist. To establish the taxonomic identity of the species of Xylopia collected in the Licuati Thicket. To use Landsat images and aerial photographs to verify the dynamics of the Licuati Thicket and associated Licuati Forest between 1958 and Research questions How many MC endemic taxa occur at the Licuati Thicket? How many woody plant species are found in the Licuati Thicket? 5

13 Is the Xylopia in the Licuati Thicket conspecific with X torrei? What size is the area covered by Licuati Thicket? Which species in Licuati Thicket do local communities use? Has the structure, distribution and size of the Licuati Thicket been changed by timber and charcoal production? 1.4 Layout of the report This report has been divided as follows. A Table of Contents (pp. ii-iii), and a List of Figures and Tables (p. iv) are followed by an Abstract (pp. I-2) and a condensed Project Proposal (pp. 3-6). A General Introduction and Historical Background are given in Chapters I and 2 respectively. Chapter 3 contains the General Materials and Methods, while Chapters 4 and 5 are presented as two independent manuscripts, both compiled in the format required by the scientific journal Bothalia. Chapter 4 deals with the Licuati Sand Thicket, and Chapter 5 with the taxonomic identity of Xylopia torrei. Appendix I is a Checklist of the Plants of the Licuati Thicket, while several miscellaneous photographs, numbered from I to 25, are presented in Appendix 2. Numbers in the text that are in bold and between square brackets refer to the photographs contained in Appendix 2. A Summary of the report is provided on pp and mainly covers the information contained in Chapters I, 2, 4 and 5. Acknowledgements, General References, and the Curriculum Vitae of the author are provided on pp

14 CHAPTER! GENERAL INTRODUCTION TO THE MAPUTALAND CENTRE OF ENDEMISM 1.1 Background Historically, the name Maputaland has been applied, broadly speaking, to the coastal plain stretching from Maputo in Mozambique, southwards to near the Mkuzi River in northern KwaZulu-Natal (Bruton & Cooper 1980). The southern border of this region has in recent times been extended further south, to the estuary of Lake St Lucia, to include a greater part of a well-demarcated biogeographical region. Maputaland has previously been known as Tongaland-a name that has fallen into disuse among the local inhabitants, and which is no longer recommended (Van Wyk & Smith 2001). In his classification of Africa into large phytogeographical regions or phytochoria, White (1983) included the area now known as Maputaland in his larger 'Tongaland-Pondoland Regional Mosaic'. Maputaland has also been considered a 'transition zone' between the northern tropical/subtropical flora and the southern temperate flora (Moll & White 1978; White 1983). This view ofmaputaland as a 'mosaic' or 'transitional zone' long obscured the importance ofmaputaland as a centre of endemism in its own right (Van Wyk & Smith 2001). When the World Wide Fund for Nature/International Conservation Union (WWF/IUCN) declared southern Mozambique and KwaZulu-Natal a 'Centre of Plant Endemism'in 1994 (Davis et al. 1994), the high levels of endemism-spread across the whole taxonomic spectrum, including both animals and plants-of the region were highlighted. Van Wyk (1994) defined the boundaries of the Maputaland Centre of Plant Endemism [MC], and also provided information on the rich biodiversity of this centre (Van Wyk 1996). 7

15 1.2 Physical description Topographically, the MC lies in a nearly flat, low-level coastal plain with a maximum elevation of about 150 m. Some extensive wetlands, including marshes, lakes and estuaries, occur within the centre. The geology of the coastal plains consists mainly of Cretaceous to Recent marine sediments, with shallow marine, littoral and coastal dune deposits (comprising the Cenozoic Maputaland Group) representing a series of marine regressive events following marine transgressions (Botha 1997; Maud & Botha 2000). The geological history of the region suggests that the current ecosystem in the area may be of recent derivation-a perception that is further exemplified by the fact that many of the endemic plant taxa occurring in the region comply with the concept of neoendemism. Neoendemism or neoendemic refer to those plant and animal taxa that occur in a restricted area and are considered to be newly derived or recently developed from earlier (possibly widespread) ancestors. This is opposed to the concept of palaeoendemism which can also refer to a taxon with a restricted global distribution, but in which case the taxon are considered the last remnant of an ancient taxon that is slowly dying out due to natural causes and which is slowly being confined to an increasingly shrinking area of distribution. An area that is characterised by neoendemic taxa such as Maputaland can, in Africa, be considered a unique environment, as Africa is more usually considered a palaeo-environment (or mosaic of palaeo-environments) characterised by palaeoendemic taxa. This is because, geologically speaking, Africa is quite old, while Maputaland is geologically young, with biological evolution (notably speciation) still very active in the region (Siebert et al. 2002). 8

16 1.3 Climate Maputaland lies within a transitional zone between the tropics to the north and the subtropical coastal conditions to the south. Summers are hot and winters cool to warm with no frost (mean temperatures: 16 C in July, 27 C in January). Schultze (1982) describes the climate of the region as warm to hot, humid and subtropical. Annual rainfall averages about mm along the coast, but progressively declines inland to as low as mm on the western plains. Relative humidity is generally high on the plains, even away from the coast (Van Wyk & Smith 2001 ). Climate plays a primary role in defining the vegetation of Maputaland. The east to west change in rainfall superimposed on the north to south and tropical to subtropical transition, together with an intimate relationship between geology, geomorphology and soils, has resulted in six main terrestrial ecozones being identified. The two easternmost ecozones-the coastal dune zone and the coastal lakes and wetland zone--developed in close relationship with the marine environment, with the dunes playing a major role in cutting off the estuaries from the sea, resulting in lakes that gradually becomes segmented to form wetlands (Siebert et al. 2002). 1.4 Flora and vegetation of Maputaland Vegetation The vegetation of Maputaland is very diverse and includes different types of grassland, bushveld, thicket, forest, and swamp vegetation. Two remarkable vegetation types, Sand Forest and Woody Grassland, are endemic to Maputaland (Moll1977, 1980). Maputaland is covered with woodland, with patches of short and tall forest usually bordered by grassland. Wild & Barbosa (1967) mapped the central area as dry, semi-deciduous 9

17 (lowland-sublitoral) forest, with woodland and savanna woodland (southeastern sublitoral) also occurring. The woodland and savanna woodland contain many extra-tropical species. Floristic composition alone is usually not enough to differentiate between the different types of woodland communities, and structure and density characteristics are usually used to help distinguish between them. The woody grass lands of Maputaland, or 'the underground forests of Africa' as White ( 1967) liked to refer to them, are characterised by the many geoxylic (often rhizomatous) suffrutices that occur there. Geoxylic suffrutices, often referred to as 'underground' or 'stunted' trees, are dwarf woody plants with annual or short-lived woody shoots sprouting from massive or extensive woody, perennial, underground axes. Common examples of geoxylic suffrutices that occur in the grass lands of Maputaland are: Ancylobotrys petersiana, Diospyros galpinii, Eugenia albanensis, E. capensis, Parinari capensis subsp. incohata and Salacia krausii. The woody grasslands are also rich in plant species endemic to the MC (Lubbe 1997) Flora and endemism The fauna and flora of the MC are predominantly of Palaeotropical (Afrotropical) derivation, and Cape floristic elements are very rare among the endemics. Although palaeoendemics are present in the region, most of the Maputaland endemics appear to be of fairly recent derivation-a perception supported by the fact that some are differentiated at the infraspecific level only, with their nearest relatives still extant. The MC is therefore outstanding specifically for its richness in neoendemics; and biological evolution, including speciation, appears to be especially active amongst both the plants and the animals in the region (Van Wyk & Smith 2001). At least vascular plant species, of which at least 230 species/infraspecific taxa are endemic or near-endemic, occur within the MC (Van Wyk 1996; Van Wyk & Smith 2001). Three genera, namely Brachychloa, Galpinia and Helichrysopsis are endemic or nearendemic to Maputaland. The plant families Asclepiadaceae, Rubiaceae, Liliaceae, 10

18 Acanthaceae, Euphorbiaceae, and Asteraceae contribute the largest numbers of endemic taxa to the region. Nearly all life forms and life history types (including saprophytes, hemiparasites, and monocarpic plants) are represented among the endemic taxa, but most of the endemics tend to be trees, shrubs, geoxylic suffrutices, lianas, forbs, geophytes and annual herbs (Van Wyk 1996). The presence of large numbers of annuals and short-lived perennials in Maputaland is noteworthy. In southern Africa, where most of the centres of endemism lie within areas that receive mainly summer rainfall, annuals are usually either rare or absent. Most of the Maputaland endemics are associated with two plant communities. Forty-two endemic taxa are associated with Sand Forest (Drypetes arguta-uvaria lucida subsp. virens forest), with 27 taxa being restricted to it, and 37 (13 restricted) are associated with the Strychnos madagascariensis-combretum molle woodland matrix (Matthews et al. 1999). The Themedeto-Salacietum woody grassland, which occur closer to the coast, is another plant community noted for its high number of MC-endemics, and 20 endemic/near-endemic taxa have been recorded from the Spirostachys africana-berchemia zeyheri closed woodland, which occur on clay soils, and not on sand as most of the other plant communities in Maputaland. Although many of the endemic plants have wide distributions within the MC, quite a number occur more locally. Some of the MC-endemics are restricted to the southern parts of Maputaland, while others are known only from the northern parts. A third group of endemics is largely confined to the Lebombo Mountains that more-or-less form the western border of the MC. Succulent endemism is largely confined to the genera Aloe and Euphorbia, and the tribe Stapelieae. Approximately 50 succulent taxa are endemic to Maputaland. 11

19 1.4.3 Forests of Maputaland White (1983) classified the forests of southern Africa into two phytochoria, the Afromontane Archipelago Regional Centre of Endemism (which extends to the northeastern African uplands and then to West-Africa) and the Tongaland-Pondoland Mosaic (which is largely restricted to the eastern coastal belt of southern Africa). White's phytochoria were not mapped at a fine enough scale to separate the Sand Forest of northern KwaZulu-Natal and southern Mozambique from the other Tongaland-Pondoland vegetation types such as Coastal Forest (Matthews et al. 1999). Low & Rebello ( 1996) divided the forests of southern Africa into three main types, namely Afromontane, Coastal and Sand Forest. Two of these, namely Coastal and Sand Forest occur within Maputaland, mosaics of which occur in some places closer to the coast. Long-vanished Coastal Forest might even be the ancient precursor of Sand Forest (W. Matthews pers. corn.). In many parts of Africa, the boundary between forest and adjacent vegetation types (usually grassland or woodland, and rarely fynbos) is very abrupt, and in nearly all cases the transition does not correspond to environmental discontinuities. The abruptness of the forest boundaries is usually ascribed to fire (Hopkins 1979; Midgley et al. 1997), an important determinant of forest edge species composition. Forest rarely bum-fires usually stop at the forest margin. In the case of the Sand Forest ofmaputaland, not only does it have distinct boundaries, but it also has a narrow zone (1-3 m wide) of nearly bare soil around the edge. There are indications that these narrow bands of bare soil may be the result of allelopathic chemicals produced by Sand Forest species, which prevents the establishment of other plants immediately adjacent to the forest. This means that Sand Forest, through allelopathic means, possibly establishes a unique environment for itself that also results in specific grassland communities neighbouring the Sand Forest. The bare zone may also play a role in the protection of Sand Forest against fire, by acting as a kind of natural firebreak (Matthews et al ). 12

20 In most parts of Africa, forests form mosaics with moist grassland and savanna(= woodland). Relictual patches of Afromontane and Coastal Forest may occasionally be associated with savanna, but this is usually when the site is a restricted refuge for the forest, and the type of savanna is usually more mesic. Sand Forest occurs mostly in a mosaic with dry savanna and thicket (Matthews et al ) Sand Forest Sand Forest occurs under drier conditions than most other forest types, and is therefore classified as tropical dry forest. According to Marker & Russel ( 1984), the presence of mixed forest in the low rainfall areas of the Eastern Cape can partly be ascribed to the presence of fog and to the water holding capacity of the deep sands, which sustain forest growth. The Sand Forests of Maputaland are also found on deep sand, and periodically experience heavy mist, especially in winter. Mist capturing by Sand Forest species can greatly increase the quantity of available water within the forest community, and therefore probably plays a role in the maintenance of Sand Forest in the relatively dry areas they occur. Research by Matthews et al. (2001) has shown that Sand Forest, when compared with Afromontane and Coastal Forest, is exceptionally rich in endemic species. These mainly neoendemic taxa are centred in the core area of the MC; that is the area south of Maputo Quissico (Mozambique) and north of Lake St Lucia in northern KwaZulu-Natal (South Africa). Sand Forest (as floristically defined) is, as far as we know, more-or-less confined to the MC. It tends to be patchily distributed within Maputaland in characteristic north-south oriented strips (Vahrmeijer 1966; Moll 1978a, 1980; Moll & White 1978; Ward 1981 ). Tinley (1971) mapped the distribution of a dry, semi-deciduous forest type (probably Sand Forest) as a thin zone stretching from northern KwaZulu-Natal up to Quissico (approximately 100 km south of Inhambane) in Mozambique. The northern limit (from Maputo-Quissico) of this dry semideciduous forest is still unclear, as Tinley noted a change in the species composition in the 13

21 area north of Maputo Bay. The term Sand Forest has been used in general (e.g. Goodman 1990; Midgley et al. 1997) and popular (e.g. Pooley 1993; Craib 1995) writings. In South Africa, Sand Forest refers to dense forests with numerous trees and shrubs (De Moor et al. 1977; Moll & White 1978; Moll 1978b; 1980), with a relatively short canopy. Different heights have been defined for Sand Forest: 6 m or higher (De Moor et al. 1997), m (Moll & White 1978; Moll 1978b, 1980), 5-13 m, and terming it 'emergent' when it reaches height of above 15 m (Ward 1981 ). Sand Forest occurs under fairly dry conditions, receiving an average of mm (Tinley 1967), or more conservatively mm (Moll & White 1978), of rain per year. It occurs on white to deep red soils (Vahrmeijer 1966; Tinley 1967; De Moor et al. 1977; Moll & White 1978; Moll1978b, 1980; Ward 1981). Sand Forest is best developed on the ancient north-south trending sand dunes of the coastal plain to which it is restricted in northern KwaZulu-Natal and the extreme south of Mozambique. Sand Forest is mainly evergreen, with a unique biological composition, including many MC plant and animal endemics (Van Wyk & Smith 2001). Because of its restricted occurrence and unusual species compliment, Sand Forest or Licmiti Forest is one of the most important plant communities in the MC (Matthews et al ), and provides rich resources for sustainable rural livelihoods (Dutton & Dutton 2000). Kirkwood & Midgley ( 1999) assessed the floristic uniqueness of the woody component of Sand Forest in relation to a range of other forest types in northern KwaZulu-Natal. Characteristic woody species of Sand Forest include Balanites maughamii, Cola greenwayi, Croton pseudopulchellus, Diallium schlechteri [18], Drypetes arguta, Hyperacanthus microphyllus [23], Newtonia hildebrandtii, Psydraxfragrantissima [17], Pteleopsis myrtifolia, Ptaeroxylon obliquum, and Uvaria lucida subsp. virens. 14

22 1.5 Forest Reserves in Mozambique Forest reserves in Mozambique are included in protected areas and are defined as a delimited territorial area, which represents natural and national patrimony, and have the role of conserving the biodiversity and endangered ecosystems and the animal and plant species contained within. Most of the Forest Reserves in Mozambique were established about 50 years ago during colonial rule, and the first objective was to protect woody species with an economical value. As part of the protected environment, the forest reserves were created to protect the vegetation types that contained species with commercial value and which were harvested intensively. The LFR, for instance, were created to protect the tree species Afzelia quanzensis [25] (or Chanfuta as it is known locally). The conservation status of most of the Forest Reserves of Mozambique is low, as most of them are degraded due a high demand for, and exploitation of, woody species for fuel, wood and timber. Some, however, are still in pristine condition as they were protected and conserved by local communities for their cultural and traditional value. Though Mozambican law makes provision for the protection of the forest reserves with three main laws, namely the Land Law, Environmental Law, and Forest and Wildlife Law, enforcement is very low, and additional measures had to be taken to control the exploitation of resources at the LFR. 15

23 1.6 The Licuati Forest Reserve [LFR] History and Location The LFR is situated about 70 km south of Maputo, close to the South African and Swaziland borders. Locally it is referred to as Licuati Forest and is also known as Sand Forest in South Africa. The major vegetation type of the reserve is Sand Forest. Licuati Forest has many rare plant and animal species, and is classified as tropical dry forest. The LFR was declared as such on 14 December 1943 by portaria 5354, to protect woody species, particularly Afzelia quanzensis [25] (Gomes e Sousa 1968). Although this reserve does not have well-defined boundaries, it is located between the Tembe and Maputo Rivers, and the Porte Henrique-Bela Vista, and Bela Vista-Catuane roads, at 26 20' Sand 32 30' E. The reserve and surrounding areas cover about ha (Mangue 1999). However, we doubt the ea ha figure (for total area covered by forest) by Mangue (1999). The name LFR has been applied for convenience, since the name Licuati refers to the dense stands of impenetrable Sand Forest, and not to the areas where most of the Afzelia quanzensis [25) is found. In most households in Maputo, firewood has been replaced by charcoal as fuel of choice, mainly because of its higher energy density and because it is cheaper to transport. In parts of the forest where resources are still abundant, the process of charcoal production consists of felling selected tree species of selected sizes. Piles of logs are then covered by soil and burned. Trees, which are preferred for charcoal and firewood are Acacia nigrescens, Albizia forbesii, Combretum hereroense, C. imberbe, Dichrostachys cinerea, Newtonia hildebrandtii and Terminalia sericea (Mangue 1999). 16

24 1.6.2 Flora and Vegetation The flora of Licmiti Forest is very diverse and includes typical sandveld species such as Pteleopsis myrtifolia, Cleistanthus schlechteri [21], Hymenocardia ulmoides and Monodora junodii, which Van Rooyen et al. ( 1981) also recorded in sandveld thickets from the Punda Milia-Pafuri-Wambiya area in the Kruger National Park. According to White ( 1980), the LFR is characterised by large numbers of species such as Afzelia quanzensis [25], Albiziaforbesii, Dialium schlechteri [18], Balanites maughamii, Spirostachys africana, Erythrophleum lasianthum, Hymenocardia ulmoides, Newtonia hildebrandtii, Terminalia sericea, Strychnos madagascariensis and S. spinosa. These are all species that are typical in ecosystems where both water and nutrient availability control primary production. Such ecosystems are characterised by episodic rainfall, separated by dry periods (Mangue 1999). Moll & White (1978) considered some of the typical Licuati Forest species (such as Balanites maughamii, Cleistanthus schlechteri [21] and Newtonia hildebrandtii) to be Zanzibar Inhambane linking species. (The Zanzibar-Inhambane Region is the coastal phytochorial mosaic extending from southern Mozambique to east Africa.). According to Matthews et al. (2001), Midgley (1993) claimed that Licmiti Forest is dominated by species with wind-dispersed seeds, whereas other Indian Ocean Coastal Forests are dominated by fleshy-fruited species. From an analysis of unpublished data on the dispersal patterns in the Licmiti Forest of the Tembe Elephant Park in northern KwaZulu Natal, it can now be shown that only about 5% of the species are definitely wind-dispersed. (This data applies to the more coastal Licuati Forest types.) Leaf attributes are generally fairly similar amongst all forest types. Licuati Forest is no exception, with plants generally evergreen, non-spinous, and leaves predominantly simple, micro or mesophyllous and entire. Semi-deciduous species do occur in Licuati Forest, but 17

25 these species are not dominant. Protruding crowns of many larger species are usually covered by epiphytes such as the wiry orchid Microcoelia exilis and various lichens including Usnea spp. According to Van Wyk (1996), the MC houses 225 endemic plant species, 30 of which are associated with, and about 20 of which are restricted to, Licuati Forest. The LFR is also home to many medicinal plant species, including Warburgia salutaris, Balanites maughamii, Tarenna species, Securidaca longipedunculata, Zanthoxylum species, Bridelia cathartica, Synaptolepis kirkii, Indigofera species, Xylotheca kraussiana, Acridocarpus natalitius var. linearifolius, Erythrophleum lasianthum, Brachylaena huillensis, Ochna spp., Garcinia livingstonei and many others. Traditional medicine is very popular, and more than 80% of the people who live in the vicinity of the LFR use local knowledge about natural resources for their healthcare. Although the knowledge and use of medicinal plants are centred or dominated by traditional healers, it not restricted only to them. Every member of the community has a reasonable knowledge of first aid treatments, particular with regards to the most common diseases like diarrhoea, fevers, snake and insect bites. The forest is, however, highly valued by the traditional healers who collect most of the medicinal plants. Their knowledge is passed on in the family after their death by the means of the spirit or medium (Mangue 1999). The utilisation and harvesting methods of the traditional healers are generally sustainable and follows the rules and regulations of the trade. Overexploitation of traditional medicinal plants is mostly the result of large scale commercialisation by urban vendors, a breakdown of taboos and the introduction of more efficient implements such as cane knives and metal axes with which to do the collecting (Cunningham 1997). The most common parts of plants collected are leaves, roots and bark. Other valuable species in the LFR are timber species such as Afzelia quanzensis [25], 18

26 Newtonia hildebrandtii and Terminalia sericea. Local houses are mostly built using products from the forest, grassland and riverine vegetation. The harvest of trees and shrubs for construction are usually selective, with durability and insect resistance two of the most important characteristics for selection. The preferred species for standing poles, and lattice and roofing structures are Terminalia sericea, Pteleopsis myrtifolia, Antidesma venosum, Croton pseudopulchel/us, Brachylaena disco/or, Balanites maughamii, Hymenocardia ulmoides, Sapium integerrimum, Gardenia volkensii, Monodora junodii, and Uvaria lucida. Walls are covered with Phragmites austra/is or Typha capensis. The most preferred thatching material are obtained from the grasses Cymbopogon excavatus, Jmperata cylindrica and Hyperthelia dissoluta (Mangue 1999). 1.7 SABONET expedition As part of a Southern African Botanical Diversity Network (SABONET) initiative, a regional expedition to the MC was conducted from 23 November to 12 December Ponta do Ouro, the Maputo Elephant Reserve and the LFR where among the localities visited. The broad objective of this regional expedition was to gather additional data on the botanical diversity of southern Mozambique for use in future projects that are related to initiatives such as transfrontier parks, sustainable medicinal plant harvesting, ecotourism, and rural development, by local botanical institutions. There is a great scarcity of basic data within the fields ofbiodiversity and conservation in Mozambique, and basic assessments of the flora of Mozambique will greatly benefit the region. At the moment it is a huge disadvantage that so little sufficient floristic data are available that can serve as a background for the establishment of proper management plans and the monitoring of developmental activities. The area is also one of the few places in sub Saharan Africa where evolutionary processes such as active speciation and the establishment of plant communities within young ecosystems can still be studied (Siebert et al. 2001). The baseline data obtained during this expedition will be of immense value to the end-users of 19

27 botanical information in both Mozambique and the rest of southern Africa. Regional and local experts who assessed the floristic richness and level of endemism of the area provided baseline floristic data for the LFR. The assessment was done in many parts of the forest officially known as the LFR. The reason why I chose to carry on with a study of the LFR was because I wanted to understand the existence of the short Sand Thicket more-or-less at the middle of the reserve, its specific composition, level of endemism, the dynamics, utilisation and the conservation of this ecosystem New record of Xylopia species The first record of a Xylopia sp. at the LFR was made during the SABONET Mozambique Expedition of December Until then, this genus was only recorded from Nampula, Zambezia and Gaza provinces. After some studies, including macromorphology, anatomy, and SEM-investigations at the University of Pretoria, and comparative studies of the holotype and other specimens from the LMA, LMU, PRE, and PRU herbaria, it was established that the species at LFR is Xylopia torrei [12,13,14], which has previously only been recorded from Chibuto (Gaza Province). The research conducted on this species will be reported elsewhere. 20

28 CHAPTER2 HISTORICAL BACKGROUND 2.1 General The LFR is situated in the Matutuine District in Maputo Province, Mozambique. Matutuine has a population of nearly inhabitants (INE 1997), and the population growth rate is estimated at about 2. 7 %. Only one ethnic group, the Ronga, inhabits the Matutuine District, but most of them are also fluent in Chizinguiri (a variant ofronga), and Zulu (Mangue 1999). The capital of Matutuine is Be la Vista. The region where the LFR is situated is called Tinonganine and consists of six main areas: Tinonganine, Djabula, Xihatho, Santaca, Tanga, and Ncuvana. The total population of Tinonganine is approximately (INE 1997). The village oftinonganine, where Mr Augusto Carmone Santaca (acting chief of the area) and his relatives and family lives, is the main village in the Tinonganine area. Most of the people from Tinonganine and its surrounds are dependant on agriculture for their subsistence and survival. The Licmiti ecosystem caters to all the needs of the local communities. It provides land for farming, timber and other building materials, wild fruit and vegetables, medicine, fuel wood, fodder and household commodities. Natural services such as nutrient cycling, the creation and maintenance of soils, acting as a source of pollinators, contributing to the regulation of the local climate, and providing sound habitat for the local communities are some of the less directly noticeable contributions of the Licmiti. It is further also a source of cultural symbols and sacred sites (Mangue 1999). 2.2 The Santaca family The first king ofmatutuine was Uangobe Tembe who lived in Kapizulu (also known as Uangobe). He had three sons called Tembe, Panhela and Maputo (all carried their father's 21

29 surname, Tembe). Tembe was in charge of the Catembe area and Panhela of the Namaacha and Changalane areas. In later years, Panhela took over a part of Swaziland and called it Dhlamini because he acquired it in the day ( dhlamini means 'eat during the day' in Mswati). Panhela then changed his surname from Tembe to Dhlamini, and his sons also adopted the Dhlamini surname. Nowadays, the Swazi's do not even know that they are related to the Tembe family. King Uangobe Tembe stayed in Kapizulu with his youngest son Maputo. After his farther passed away, Maputo took over the responsibility ofuangobe, and was crowned King of Matutuine (Catuane, Machangulo, Inhaca, Catembe, and some other areas). The new king renamed his land and called it Maputo. Although Maputo was the youngest of the three brothers, his political power was much stronger than either Tembe's or Panhela's. Maputo moved from Kapizulu to the Salamanga area, also called Madlhadlhane. Maputo had many sons, of which the most renowned were Salamanga and Makasane. The Matutuine area was renamed as Tembe, after the family name, but later became known under its original name again. Makasane, one of the sons ofmaputo, had two children, a son called Muwai, and a daughter called Diwandinwa. Muwai took two wives, and had three sons; Musongue and Ngohika from the first wife, and Makasanhe from the second. Muwai named his third son after his father, Makasane, but out of respect for his father, he altered it to Makasan(h)e. Muwai passed away before his own father, and did not get the opportunity to rule the land. When his grandfather Makasane died, Musongue became king. Musongue's mother's labola included large herds of cattle, which made her and her family very respected among the local people. This made Musongue a very important person, and a suitable heir. He got married and had one son, Gwanaze. His brother, Ngohika, also had a son-santaca I. When the Portuguese arrived in Mozambique, Gwanaze ran away to South Africa where he resettled close to the border with Mozambique in the area then called Manguzi, now 22

30 Gwanaze. Gwanaze took responsibility for the area close to Ponto do Ouro on the South African side. A member of the Madlhadlhane family, Powowo, took over from Gwanaze, and became the Chief of the area. Since Gwanaze had fled from Mozambique, Santaca I (son ofngohika) became the new ruler. Santaca I had no sons, and when he passed away, the power had to be passed to one of his sisters' sons. Santaca, son ofngohika, had two sisters; the oldest only had two daughters, but the youngest had three sons, including Santaca 11, the father of the current Acting Chief Augusto Santaca and his brothers. Santaca II was a teacher and a medical assistant, the first of both professions in the Santaca area, and was well respected there. During the colonial regime, Santaca 11 adopted the name Jorge Oliveira Santaca, and when he became chief of the area, his mother received a large labola of cattle, which made them well respected among the local people, and therefore, even though he was not the son of a chief, he became the new chief of the region. After Mozambique gained independence from colonial rule, the traditional power was no longer recognised, and Jorge Santaca moved to South Africa where he stayed until his death in November He left behind three wives and many sons. They live in both Mozambique and South Africa, close to the border between these two countries. Jorge Santaca's body was burned in the Sacred Forest (see discussion below). Two sons from Mozambique, Chief Augusto and his brother Edmundo Santaca, attended the funeral of Jorge Santaca. They are planning to move all their family in South Africa back to Mozambique to choose the new chief. The new chief will be one of the brothers of Santaca 11. If the chosen brother declines the chiefdom, one of the sons could take over by family consensus. There are eight brothers, two staying in Tinonganine, one in Chimoio, two in Maputo, and three in South Africa. All must be present at the family meeting before a new chief can be chosen, as the government of Mozambique has now recognised traditional power in the country. Mr. Augusto Carmone Santaca is presently the acting chief of the area where part of the 23

31 Licmlti Forest is situated, and is also the Secretary of the Frelimo Party, by Frelimo nomination. 2.3 Interviews with Chief Santaca (Conducted by author on 10 October 2002) When I did my fieldwork at the LFR, I stayed at Tinonganine and had a great opportunity to interview the chief and his youngest brother, Mr. Edmundo Ti6filo Santaca. Chief Augusto Carmone Santaca was born in the area called Santaca and is 62 years old. He has been a driver and mechanic for most of his life, but is presently a farmer. He was involved with the Blanchard project, working at the Maputo Elefante Reserve. In 2000 he worked at the Fabrica de Cimentos de Mo9ambique. Mr. Edmundo Santaca was also born in Santaca, and is 55 years old. He has been a farmer since 1968, and had a small grocery shop in Tinonganine village, which he had to close during the war. He returned to his original profession after the war. The LFR lies within the boundaries of the land controlled by Chief Santaca and his family, who therefore know a lot about the natural and cultural history of the area. According to Chief Santaca, the name LFR was applied only for the sake of convenience, as the name Licmiti refers to Sand Thicket, where very few Afzelia quanzensis [25] (Chanfuta) are found. There are no official boundaries for the LFR, but to protect Chanfuta, it is necessary that new boundaries for the reserve be defined-boundaries that exclude the Licmiti Thicket, and specifically protect the areas where Chanfuta is found. The chief and his brother claims that the elephants and other big game that belong to the Maputo Elephant Reserve migrated long ago from the Licmiti Thicket to area in which this reserve is now located. They believe, that, in order to develop and expand the ecotourism potential of the LFR, it is necessary that the government reintroduce some animals that used to occur there, e.g. jackal, impala, kudu and other game, to the reserve. It is also important 24

32 that the habitats of the bird species that are endemic to the region be conserved. Staff from the Ministry of Agriculture and Rural Development put pole fences around the thicket section of the LFR to mark the areas of different Ndunas or 'regulados'. This work was terminated even before the war. Most of these areas belong to the Santaca family, but within the boundaries of the bigger area, there are smaller areas that belong to different Ndunas. The fenced area demarcates the boundary between Kwane and Djabula. 2.4 Medicinal Plants The use of traditional medicine is very popular within the area, and many local people have a good working knowledge about local medicinal plants. The people can collect medicinal plants from the forests without consulting with a traditional healer. Knowledge has been passed down over many generations and is used to cure almost all of the diseases commonly occurring in the area. The ordinary people do consult traditional healers when the symptoms persist or when they believe that the diseases are linked to spiritual problems. It is believed that the traditional healer has the power to cast out the disease-causing spirits and to prescribe medicine for the patient. To the local communities, the traditional healer is a medical doctor and spiritual leader all in one, someone who can cure them from all sicknesses and diseases. Both tall forest and Licmiti Thicket are very important sources of medicinal plants. Tall forest, however, is the part that is subjected to the most use. As a result, the Licmiti Thicket is not under much pressure from medicinal plant collectors. 2.5 The Sacred Forest The Sacred Forest was the home of the first chief and is now the cemetery of the Santaca family. The fact that this part of the forest has this special status, and the protection that goes with it, has resulted in the conservation of some important species that occur in the area. The local community has respect for this forest, and are not allowed to enter it without permission 25

33 from the Chief. The Sacred Forest is also the site where the Santaca family asks for rain, good harvests and good luck for the community. This is known as the Marula Ceremony, and marula wine is made and prayers for a prosperous year are prayed. Chief Santaca wants to open up the Sacred Forest for ecotourism purposes, as he believes that all traditional ceremonies should be done and known by all visitors. 'Everyone who is interested to know us and to understand our culture should know the history', says the Chief. I had the opportunity to attend a traditional ceremony where the Chiefs brother presented a chicken and wine to introduce us to their ancestors. They believe that the ancestors should be informed about all visitors to the Santaca area, and about the objective of such a visit, to ensure good I uck and smooth working operations. 2.6 Utilisation Chief Santaca indicated that the areas of the reserve that is subject to the most use is those parts where most of the Afzelia quanzensis (25] and other woody plants such as Combretum spp., Terminalia sericea, Sclerocarya caffra, Albizia species, Newtonia hildebrantii, Ekebergia capensis, Acacia spp. etc.-i.e. the tall forest and woodland-occur. Outsiders, mainly from Maputo, have also utilised these parts of the forest, mainly to exploit Afzelia quanzensis (25] (Chanfuta). The harvesters from Maputo return there via Goba, as they are well aware of the fact that the Directorate of Forestry and Wild Life check all vehicles for illegal timber at Porto Henrique. They return via Moamba roads to avoid this enforcement. The local communities mainly use the tall forest to selectively collect building material and firewood. During the colonial times, law enforcement was very effective, and the local community was involved via the chief of the area. The Forest Reserve was placed under the protection and 26

34 control of the local chief, and he had to be consulted before any trees could be harvested. The chief was authorised to allocate trees and would then inform the government of the allowed areas and quantities of wood that can be harvested. Nowadays, the lack of law enforcement, together with a decrease in the local chiefs authority, has resulted in the exploitation and overutilisation of the forest as the demand for timber increased. An attempt has been made by Chief Santaca to regain some control of the forest and the harvesting. The local communities now inform the chief when they find evidence of unauthorised timber cutting and charcoal production. The chief then informs the Department of Agriculture and Rural Development in Bela Vista, who then send officials to confront the illegal harvesters and apply a penalty according to the wood or charcoal quantity, and the product is confiscated. The LFR is also used for hunting. Hunters have been known to use fire to try driving the game out of the forests, but this strategy mostly failed, as the animals tend to run deeper into the thicket. Local hunters have been advised to stop this practice because the destruction of many useful plants (e.g. medicinal or edible plants) by the fires. The fire is also dangerous for the people who live close to the forest. The local hunters now use traditional traps, namely holes covered with grass or snares. The animals fall into the traps or catch their heads in the nooses, and the hunters then come by later to kill them. The local communities also use parts of the reserve to grow crops and build their houses Economic and social development During colonial times, until the start of the war, the main economic activities of the local people were stock farming and growing crops. This was called' Junta do Povoamento' (a project created by the Directorate of Agriculture oflouren<;o Marques during colonial times). The main objective of this project was to network all the farmers so that they work collectively, and 5 ha were allocated to the local people for every 10 ha allocated to the community. 27

35 All equipment was shared and donated by the Project. It was used in benefit of all the farmers. As a result, all farmers could produce several crops, and surplus harvests were sold at the markets in Maputo. In this way, the local people had enough food and some money for other expenses. Each farmer also received five cows from the Project. Surplus milk was also sold at the markets of Maputo. Veterinary Services, which also inspected the livestock, provided a vehicle to transport the milk to Maputo. Currently, the local rural people face many difficulties. They no longer have their own farms, and they don't own any livestock. The quality of their lives has gone down since 'Junta do Povoamento ', and they suffer from hunger, and lack basic facilities such as health care, educational facilities, and clean water. As a top priority, the people would like to see the good things of the past return so that everyone can again live in dignity and health, in a community where all benefit from communal farming efforts. Only one school currently operates at Tinonganine, and the children have to walk long distances to get there. This school offers only basic education (primary school up to grade 5). The people would like to see more schools in other locations such as Santaca and Djabula, to minimise the problem. Most of the kids of the area do not proceed past primary school, as their parents do not have the funds to send them to other schools in Bela Vista, Salamanga, Boane or Maputo. The schools in Salamanga and Bela Vista in any case offer education only up to grade 7, which make them still only primary schools. The local communities would like to have some secondary schools in the area so that their children can progress further in their education. They would also like to see more teachers and equipment at the schools. During colonial times, a medical doctor from Maputo came to visit once a week, and one medical assistant has permanently been on standby with all the medicines and medications. Now there is only one small clinic, with no doctor, and most of the time, without any medications. There is only one sanitary worker who visits once a week. In case of sickness, 28

36 the people have to seek private transport to Bela Vista or Maputo to see the doctor. As there is most of the time no transport available, people have to walk with the sick person to Be la Vista. Public telephones and mail services were available during colonial times, but this infrastructure was destroyed during the war. Now, people have to travel to Bela Vista whenever someone has to make an urgent telephone call. Communication with the outside world is very difficult, and the people hope that this situation would improve in the future. There also is no electricity. Mr Edmundo Santaca was informed that there was plans to built a dam on the Xuxa and Madiwa rivers in order to divert the water to Tembe River. This could avoid floods which damage crops, and this project is therefore very important for all the communities, as they will not only benefit from the dammed water during dry seasons, but they and their livelihoods would also not be threatened by the flooding of the Maputo river during the wet seasons. We would like to urge the government of Mozambique to proceed with this project. 29

37 CHAPTER3 MATERIAL & METHODS Though summaries of the materials and methods used in the investigations are provided in the chapters written up as articles according to the format and specifications of Bothalia (Chapters 4 & 5), a more complete description of each of the methods are provided in this chapter. Where applicable, complete references to the original accounts of the methods are given. Descriptions of the methods used are presented under headings referring to the various sources of evidence (leaf anatomy, SEM work, etc.) that were investigated. 3.1 Visit to the study area A visit to the study area was planned and executed to meet and inform the local authorities in the area (Tinonganine and surrounds), as well as the administrative authorities in Bel a Vista, about the study we were planning to conduct, its objectives and expected outcomes. To get to know and to define the boundaries of the study area-namely the Licmlti Thicket, a literature review (including reports and Flora's) were done. Landsat images was obtained and used to define and demarcate the specific areas within the Licmiti Forest Reserve that was to be used in comparative studies. The core area of this study is called the Sand or Licmlti Thicket, a very close and impenetrable vegetation type in more-or-less the centre of the LFR. 3.2 Fieldwork A collecting trip to the Licmiti Thicket, LFR (26 20' S and 32 30' E), to collect plant specimens was undertaken. Specimens were collected, pressed, dried, mounted, and sent to the H.G.W.J. Schweickerdt Herbarium (PRU), University of Pretoria, for final identification. 30

38 Some of the leaves of the new record of Xylopia spp. we found in the Thicket were placed immediately in formalin-acetic-acid-alcohol (F AA), for anatomical and morphological studies. 3.3 Herbarium studies To gain clarification on the identity of some of the species collected at the Licuati Thicket, a herbarium study was carried out using the Flora Zambesiaca, Flora de Mor;ambique, Flora of Southern Africa, dichotomous keys and scientifically updated specimens at the National Herbarium, Pretoria (PRE), the H.G.W.J. Schweickerdt Herbarium (PRU), and the LMA and LMU Herbaria in Maputo, Mozambique. Leaves of Xylopia parviflora, X odoratissima and X tomentosa, were collected from voucher specimens of different collectors from Mozambique, South Africa, Zambia and Zimbabwe. One leaf of each species was rehidrated and placed in F AA for anatomical studies, while another was saved for further macromorphological studies using Scanning Electron Microscope (SEM) technology. The same was done with Xylopia sp., collected in Licuati Thicket. 3.4 Checklist of the Licuati Thicket An annotated checklist of the plants of the Licuati Thicket was compiled using the specimens collected, preexisting checklists, and plant collections in the LMA, LMU, PRU and PRE Herbaria. Flora's, monographs and other relevant literature were also consulted in the construction of the checklist. Selected characteristics, such as growth form, fruit type, duration of flowering, endemic status, and medicinal usage were added to the checklist. 3.5 Dynamics of the Licuati Thicket The dynamics of the Licuati Sand Thicket was verified by comparing aerial photographs 31

39 taken in 1958, 1982, and 1989, to study the changes in the structure of the vegetation over a period of 30 years. Maps showing the Licmlti Forest, and a topographic map were also used in the study. The coordinates that were taken during the collection trip were used to map the Licmiti Thicket. The map was produced using the GIS program Arc View. 3.6 Historical information and current utilisation of the Licuati Thicket Interviews were conducted with Chief Augusto Santaca, his brother Mr Edmundo Santaca, and one traditional healer Mr Dinis Nguenha in the Tinonganine district to gather information on the historical and current utilisation of the Licmiti Thicket. Interviewees were asked the following 12 questions: Where is the official boundary of the Licmiti Forest Reserve? What is the local name for the Licmiti Short Forest (Thicket)? Why is there a fence around part of this short Licmiti Forest? For what purposes and how do the local people use the tall and short Licmiti Forest respectively? What do the local people think of the felling of trees in the forest and thicket? Is there any law enforcement against illegal felling of trees, and does anyone give permission for this activity? Are there any penalties when illegal cutting of timber takes place? What are the local people's feelings about the Sand Thicket? Should it be conserved? What is the meaning of the Sacred Forest? Do they know that the Licmiti Short Forest (Thicket) of the Licuati Forest Reserve occurs only in this area and that it is unique? Are there many people who use medicinal plants for their health care? Do they consult traditional healers before they go into the forest to harvest the medicinal plants? What are the main economic activities in the area, and what do they think about rural development initiatives? 32

40 3. 7 Studies on Xylopia: Leaf Anatomy The leaf anatomy of the Xylopia species found within the Licuati Thicket of the Licuati Forest Reserve was studied using light microscopy (LM). The material was prepared as follows: Dehydration and imbedding of leaf material (Adapted from Feder & O'Brien (1968)) Pieces of leaf lamina (3 mm x 3 mm) were cut from material fixed and preserved in F AA using a sharp razor blade. The pieces were cut to either include a part of the lamina between the midrib and the leaf margin, or to include midrib bordered by small parts of lamina. Pieces were cut± 1/3 of the length of the leaf from the leaf base. The pieces of leaf material were washed using 50% ethanol (EtOH), dehydrated using a graded alcohol series, and finally infiltrated with glycol methacrylate (GMA) * 1: 2 x 10 minutes in 50% EtOH 2 x 30 minutes in 70% EtOH 2 x 3 hours in 96% EtOH 2 x 4 hours in 100% EtOH 2 x 4 hours in 100% propanol 2 x 4 hours in 100% butanol 2 x 15 days in GMA in a dark cabinet After infiltration with GMA, the pieces of leaf material were transferred to marked gelatin capsules filled with new (clean) GMA. Capsules were heated in an oven at 50 C until the GMA has been completely polymerised (after three to four days). Ultra-thin section (± 4 J.Lm) was made using a Leica 2045 ultra-microtome fitted with a glass knife. Sections were mounted on microscope slides, double-stained with the periodic acid-schiff s (PAS) reaction and toluidine blue*2, and viewed using a light microscope. 33

41 * 1 The GMA monomer mixture was prepared according to the following recipe: To prepare 200 ml ofgma 188 ml 2-hydroxyethyl methacrylate (HEMA, GMA) (Electron Microscopical Sciences, Cat. No , Premier Technologies) 12 ml polyethylene glycol g benzoyl peroxide The mixture was mixed very slowly with a magnetic agitator for one hour or longer until the catalyst (benzoyl peroxide) was completely dissolved. It was then filtrated using a Whatman No. 1 filter paper, and placed under a soft vacuum for approximately one hour. *2 Staining of plant material using the PAS (periodic acid-schiffs) reaction and toluidine blue: Sections of plant material were stained as follows: Microscope slides with uncovered sections were placed in a saturated solution of DNPH (2-4 dinitrophenylhidrasine )*3 for at least 30 minutes to neutralise the aldehydes present in the material from the F AA in which it was preserved and fixed. After 30 minutes, the material was rinsed in flowing tap water for approximately 10 minutes. It was then placed in a 1% solution of periodic acid for exactly 10 minutes. (To prepare a fresh solution of periodic acid: dissolve 1 g of periodic acid in 100 ml distilled water.) The material was rinsed for 5 minutes in flowing tap water after which the microscope slides were dried on a hot plate. The microscope slides were placed in Schiffs reagent for 30 minutes after which they were again rinsed in flowing tap water for minutes. After a final rinse in distilled water, the material was stained in toluidine blue for 30 seconds, after which it was rinsed with flowing tap water until the surrounding material was clear. The microscope slides were again dried on a hot plate, and the material mounted permanently with entallen. 34

42 *3 Preparation of DNPH solutions: To make up a 200 ml solution of saturated DNPH: 0.6 g-1.0 g DNPH 30 ml acetic acid 170 ml distilled water Mix ingredients with a magnetic agitator (for several hours, or even overnight) until a saturated solution ofdnph has formed. Filtrate mixture with a Whatman No. 1 filter paper to get rid of excess, non-dissolved, DNPH. 3.8 Leaf macromorphology (SEM studies) Pieces of dried leaf material (3 mm x 3 mm) of Xylopia torrei [ 12, 13, 14}, X parviflora, X odoratissima and X tomentosa, were mounted on SEM buttons with plastic bands and liquid silver. After being sputter-coated with gold, the SEM buttons with gold plated samples were loaded into a JEOL 640 SEM to study the differences between the leaf surfaces of the five species. 35

43 CHAPTER4 THE LICUATI SAND THICKET: A UNIQUE VEGETATION TYPE OF SOUTHERN MOZAMBIQUE* *Manuscript follows format of Bothalia 36

44 Licuati Sand Thicket: A unique vegetation type from southern Mozambique S. Izidine 1, A.E. van Wyk 2, & S.J. Siebert 3!.Department of Botany, National Institute of Agriculture Research, P.O.Box 3658, Maputo 8, Mozambique 2. Department of Botany, University of Pretoria, Pretoria 0002, South Africa 3. National Botanical Institute, Private Bag X 101, Pretoria 0001, South Africa KEYWORDS: Ecology, endemism, ethnobotany, flora, indigenous knowledge systems, Licmiti Thicket, Mozambique, Sand Forest, vegetation structure. ABSTRACT Licmiti Forest (usually referred to as Sand Forest in South Africa) is a very distinctive vegetation type, not only because of its unique combination of rare and endemic species, but also because it has a restricted global distribution. This dry tropical type of forest only occurs in southern Mozambique and northern KwaZulu-Natal, South Africa, where it is best developed on the ancient north-south trending dune cordons of the coastal plain of Maputaland. It is also one of the most important plant communities of the Maputaland Centre of Plant Endemism. Previously a distinction has been made between so-called tall and short Sand (Licmiti) Forest. Because of its vegetation structure, floristic composition, and high levels of endemism, and based on Edwards' structural classification of vegetation, we propose the name Licmiti Thicket for the short, dense form of short Sand Forest that covers about ha in the Licmiti Forest Reserve, southern Mozambique. The aim of this study was to gather information on the floristic composition, utilisation and conservation of Licmiti Thicket that will aid in the development of a vegetation dynamics model that, in turn, can be used to guide reserve management, land-use planning, and sustainable utilisation programmes to conserve this rich, endemic vegetation type for future generations. This will only be effective if the indigenous knowledge systems of the area is incorporated and used in conjunction with official law enforcement. 37

45 INTRODUCTION There exists, within the boundaries of the Maputaland Centre of Endemism [MC] (Van Wyk & Smith 2001), a vegetation type unique to this region, usually called Sand Forest in South Africa and Licuati Forest in Mozambique. Up to now, no formal distinction was made between what was informally called 'short Sand Forest' and 'tall Sand Forest', or 'short Licmiti Forest' and 'tall Licmiti Forest'. After a close investigation of the short Licmiti Forest (or what we would like to call 'Sand Thicket') of the Licmiti Forest Reserve [LFR] in Mozambique (Figure 1 ), it was decided that this particular vegetation type is structurally clearly a thicket, not a forest, and therefore merits a classification separate from tall Sand or Licmiti Forest. In his structural classification of vegetation, Edwards (1983) used specific combinations of growth form, cover and height attributes to distinguish among different vegetation types. He defined thicket as 10% tree cover with a 5-10 m height, and/or shrub cover with a height of 2-5 m. The dense thicket of the study area, the LFR in southern Mozambique, corresponds closely with this definition, and it is therefore justified to differentiate between Licmiti Forest and Licmiti Thicket. To further justify this separate classification of two vegetation types where up to now only one vegetation type, but with two height classes, were recognised, interviews were held with the local inhabitants, specifically with the prominent family of the region as represented by Acting Chief Augusto Carmone Santaca and his brother Edmundo Ti6filo Santaca. According to the Santaca brothers (ages 62 and 55, respectively), the Licuati Thicket has looked like it does today since the time that they were children. They grew up in the area and often played in the Licuati Thicket, and have never seen any tall trees in that area of the reserve. They say that the Licuati Thicket has always been unique, and that it only occurs in a specific part of the region, and does not have the structure it has as a result of over harvesting of tall trees. Rather, the Licuati Thicket has grown and established itself there with the characteristics it has today. The boundaries of the Licuati Thicket [9], are Manhihane, which belongs to Djabula Nduna, and Maduvula, which belongs to Catuane Nduna. The Djabula Nduna is part of the Santaca 38

46 tribe, while Catuane is part of the Kapizulu tribe. A small part of K wane, and ea. 2 km of Poxane also form part of the Licuati Thicket. Most of the LFR is part ofdjabula, and is therefore under the control of the Santaca family. Licmiti means 'very dense and impenetrable thicket' in the Ronga language, and the name is used to refer to the 'short forest' or thicket. The study area's thicket vegetation is the real Licmiti, and not the whole Forest Reserve, even though the LFR was proclaimed to protect Afzelia quanzensis (known locally as Chanfuta), which mostly does not occur in the Sand Thicket (the real 'Licmiti'), but rather in the surrounding Sand Forest (also called Licmiti Forest). The two main patches of thicket within the LFR together covers about ha (Figure 2), and contain many MC endemics/near-endemics such as Strophanthus luteolus, Secamone delagoensis, Dovyalis longispina, Acridocarpus natalitius var. linearifolius, Combretum mkuzense [15], Croton steenkampianus, Erytroxylum delagoensis, Ochna barbosae [24], and medicinal plants such as Manilkara disco/or, Landolphia kirkii, Synaptolepis kirkii, Oxyanthus latifolius [22], Ochna natalitia, Psidrax locuples [16], Croton pseudopulchellus, and Cassytha filiformis. These endemic and medicinal plants that occur within the thicket are well protected, as most of the local people collect medicinal plants in the surrounding parts of the LFR, where species such as Afzelia quanzensis [25], Newtonia hildebrandtii, Balanites maughami, Combretum mkuzense [15], and other forest trees, occur. In contrast to the Licmiti Forest, the Licmiti Thicket contains few of these species, and then only in relatively small numbers, giving even more impetus to the decision to differentiate between Licmiti Forest and Thicket, as it alludes to a difference in the floristic composition of the two vegetation types. 39

47 ~' ttji ) J, I J ' 3:)' 32"15' 32"3:)' 32"45' 40 0 ~~~--~~~~~ MAJOR TO'NNS /)./.PRIMARY ROADS MAJOR RIVERS LAKES CUT LINES LICUAll THICKET LICUAll FOREST RESERVE 1\1 COUNTRY BOUNDARIES 40 Ki lom eters N A Figure 1. Map showing the location of the Licuati Forest Reserve (LFR), southern Mozambique. 40

48 32" 26 20' 26 25' 26 30' 26 35' 32" 20' 32" 25' 32"30' 32"35' Kilometers N._cUT LINES ~IMARY ROADS JORRIVERS UCUATI THICKET N A Figure 2. Map showing the Licuati Thicket, Licuati Forest Reserve, southern Mozambique. 41

49 The Licmiti Thicket is also home to many small mammals, birds and insects, some of which are endemic to the Maputaland region. Although the protection of the Licmiti Thicket has not been a priority to the local people, the importance of this vegetation type is slowly being realised. The thicket do provide some medicinal plants, building material, firewood and honey, but the locals prefer to utilise the tall forest, as it is not only closer to their homes, but also easier to enter, and to collect resources and hunt in. The thicket is much more impenetrable than the forest and therefore more of a hindrance than a help during everyday activities such as those mentioned above. Due to its inaccessibility, the thicket acts as a natural refuge for small wildlife, therefore not only contributing to the protection of endemic and medicinal plants, but also to the conservation of these small animals. Chief Santaca said that he would like to see the Licuati Thicket preserved for future generations. He also said that this part of the LFR has always been the same, and that, although the local people do not know about its uniqueness and high levels of endemism, they have always had a lot of respect for this area. Now that Chief Santaca has been informed about the richness and uniqueness of the Licuati, he has become sensitised and has agreed that the area should not be exploited beyond its endurance, but rather be opened up for ecotourism, which is less destructive than many other of the possible uses to which the forest and thicket can be put. Santaca believes that many people in the world would like to see this unique vegetation type with its endemic flora and fauna, and he felt that the local community was lucky to have inadvertently conserved the Licuati for such a long time, and that they were fortunate to have all that richness and beauty right there in the Santaca district. 42

50 MATERIAL AND METHODS Checklist of the flora of the Licuati Thicket An annotated checklist (Appendix 1) of the plants of the Licuati Thicket has been compiled. Characteristics such as growth form, fruit type, duration, endemic status, common name, and medicinal uses have been included in the checklist. Voucher specimens were collected, and all local common names and medicinal uses, given by a traditional healer in the area, Mr Dinis Nguenha, were noted. Voucher specimens are housed in the H.G.W.J. Schweikerdt Herbarium (PRU) of the University of Pretoria. Dynamics of the Licuati Thicket The dynamics of the Licuati Thicket was verified by examining aerial photographs taken in 1958, 1982, and 1989, to gain an understanding of the changes in the structure of the vegetation over a period of about 30 years. Maps of the Licuati Forest and Licuati Thicket respectively, were produced using the GIS program Arc View, and Landsat images for the area, supplemented by 1: topographic maps. Historical information and present utilisation of the Licuati Sand Thicket To gather information about the historical and current utilisation of the Licuati Thicket, interviews were held with the acting chief of the area, Mr Augusto Carmone Santaca, his brother Mr Edmundo Ti6filo Santaca, and with one of the traditional healers that operates around the village oftinonganine and the Licuati Forest Reserve, Mr Dinis Nguenha. 43

51 RESULTS AND DISCUSSION Floristics of the Licuati Thicket, Licuati Forest, and the greater MC The vascular plant flora of the Licmiti Thicket comprises about 113 indigenous species (Table 1) and makes up over 51% of the vascular plant diversity in the Licmiti Forest Reserve, which contain, in addition to Licuati Thicket, also Licuati Forest vegetation. Previously a total number of 216 species have been recorded for the LFR (Siebert et al. 2002). In terms of the floristic diversity of the larger southern part of Mozambique (including the LFR and the Maputo Elephant Reserve), the Licuati Thicket contains about 37% of the total number of species occurring within this larger region. Ferns (comprising two species belonging to two separate genera and two different families) make up about 4.5% of the total flora of the Licuati Thicket. Monocots (from seven families, eight genera, and ten species) make up about 8.8% of the total flora (Table 1), while the dicotyledons (35 families, 75 genera and 101 species) make up the bulk (89%) of the Licuati Thicket flora. Both pterophytes and monocotyledons have a very low diversity at the species level (Table 1 ), indicating that these groups are not well adapted to the thicket conditions. Table 1. Number of families, genera and species of Pterophyta, Monocotyledonae and Dicotyledonae recorded from the Licuati Thicket, LFR. Pterophyta Monocotyledonae Dicotyledonae No. % No. % No. % Total Families Genera Species

52 The families of flowering plants are very well represented in the Licuati Thicket with the Rubiaceae (coffee family) contributing the most species and genera (Table 2). Twelve families each contribute more than 1% of the total number of species, and collectively account for about 58% of the total number of species. ' Table 2. Synopsis of families that contribute 1 /o or more of the total number in species of the Licuati Ticket, LFR. Family Genera Species No. o/o of total No. o/o of total Rubiaceae Euphorbiaceae Fabaceae Anonnaceae Celastraceae Asteraceae Capparaceae Sapotaceae Asclepiaceae Ochnaceae Flacourtiaceae Poaceae Amaranthaceae According to Van Wyk & Smith (2001) the six families that contributes the largest numbers of endemic/near-endemic species and infraspecific taxa in the MC are: Asclepiadaceae, Euphorbiaceae, Rubiaceae, Liliaceae (in a broad sense), Acanthaceae, and Asteraceae. This differs from what we found for the Licuati Thicket, where the three families with the largest number of endemics are the Rubiaceae, Fabaceae, and Celastraceae (Table 3), indicating that there are variations in the distribution patterns of the endemic flora within the different 45

53 vegetation types of the MC, and that there is a need for further research on this topic. The Rubiaceae contributes eight species or 26.6 % to the total number of Licmiti Thicket endemics (Table 3). Still looking at the Rubiaceae; in terms of endemism, the Rubiaceae contributes a relatively higher number of endemics to the total flora in the Licmiti Thicket, than it does in the Maputo Elephant Reserve (Table 3), where Licmiti Forest are more prominent. In the Maputo Elephant Reserve, the Rubiaceae only contributes five endemic species, or 13.8% to the total MC endemic flora of this reserve (Siebert et al. 2001). The flora oflicmiti Thicket contains 85 small trees and shrubs, of which 15% belong to the Rubiaceae family (Table 4). Van Wyk (1996) and Van Wyk & Smith (2001) estimates that about or more species of vascular plants occur in the MC. Of these, 230 species or subspecific taxa are endemic or near-endemic to the region. The Licmiti Forest contains ea. 30 species, or 13%, and the Maputo Elephant Reserve 36, or 16%, of the total number of endemic specific or infraspecific taxa in the MC. 46

54 Table 3. Synopsis of families that contribute 1 /o or more of the total number of MC endemics in the Licuati Thicket, as compared to the figures for the Maputo Elephant Reserve. Licmiti Sand Thicket Maputo Elephant Reserve Family Sp. % Sp. % Rubiaceae Fabaceae Celastraceae Apocynaceae Euphorbiaceae Asteraceae Asclepiadaceae Poaceae Anonnaceae Ochnaceae Table 4. Families, which contribute five or more small tree or shrub species to the flora of the Licuati Thicket. Number of small trees and shrubs Family No. % Rubiaceae Euphorbiaceae Anonnaceae 6 7 Fabaceae Apocynaceae Celastraceae

55 Vegetation dynamics of Licuati Thicket and Licuati Forest Based on the examination of aerial photographs taken in 1958, 1982, and 1989, we found that no significant changes in the structure and size of the Licuati Thicket and associated Licuati Forest took place during the past 30 years. This indicates that the Licuati Thicket [9] is very stable, and has not been subjected to intensive over exploitation that might have had a large influence on the structure and overall size of this vegetation type. The Licuati Thicket is characterised by the shrubs Acridocarpus natalitius subsp. linearifolius, Artabotrys monteroi, Monodorajunodii, Uvaria lucida, Dialium schlechteri [18], Hymenocardia ulmoides, and Warneckea sousae. This species composition contrasts with that of the forest edges, which are subject to more utilisation, and where tall trees such as Afzelia quanzensis [25], Newtonia hildebrandtii, Pseudobersama mossambicensis, and Albizia adianthifo/ia occur. The thicket edges and a small portion of the centre of the thicket in the LFR are subject to some disturbances, mostly fire and some tree felling. As a result of human activity in the area, fire occurrence has increased in frequency recently, and vegetation are sometimes burned during times when it would not naturally and on a regular basis, be exposed to fire. These unnatural burning episodes pose a threat to the natural recruiting processes and have been shown to have a destructive impact on the Licuati Thicket. Many of these Licuati Thicket patches and edges that are negatively impacted by unnatural burning regimes do not regenerate at all. On aerial photographs of the Licuati Thicket taken in 1989, the extent and damage caused by a fire that has occurred years prior to the taking of the photograph was still very clear. This suggests that historical plant dynamics of the Licuati Thicket might have been a major determinant of the occurrence and extent of the present day patches of this vegetation type (Izidine et al. 2002). Edwards ( 1983 ), in his structural classification of vegetation, defined a thicket as a vegetation type where there is a % tree crown cover of 2-10 m in height and > 10% shrub cover of 1-5 m in height. He defined a forest, on the other hand, as a vegetation type where there is % tree crown cover of2-20 m in height and< 10% shrub cover of 1-5 m in height. 48

56 The Licuati Thicket corresponds closely to Edwards' definition of a thicket. It is a distinct vegetation type that is relatively impenetrable and has a very specific species composition. Although Licuati Forest and Licmiti Thicket have many plant species in common, the Thicket is characterised by species such as Cleistanthus schlechteri [21], Croton pseudopulchellus, Hyperacanthus microphyllus [23], Oxyanthus latifolius [22], Warneckea sousae, Psydrax locuples [16], P. fragrantissima [17], Tricalysia junodii and Xylopia torrei [12, 13, 14]. The historical information provided by Chief Santaca is very precise in that the Licmiti Thicket has remained stable since at least the time that he and his brother have been children, and that no change in the structure has occurred for at least the past years. These personal testimonies of people that have lived here all their lives refutes the opinion expressed by some ecologists that Licmiti Thicket is just Licwiti Forest with all the tall trees chopped down. Utilisation and exploitation of the Licuati Forest Reserve The Licwiti Forest Reserve is mainly used for grazing purposes for the cattle herds of the local communities and for natural resource based industries such as the production of palm wine, reed harvesting and charcoal production. Agricultural activities take place mostly on the fertile clayey alluvium soils of the floodplains of some of the larger rivers such as the Tembe, Maputo and Futi. The main crops produced in the area are peanuts, beans, maize, cassava, vegetables, sweet potatoes, mangoes, and paw-paws. Some woody native species are also selectively utilised for timber, building materials, and fuel wood. In the case of timber, the most valuable species that occur in the LFR is Afzelia quanzensis [25] (Chanfuta), Newtonia hildebrandtii and Terminalia sericea. Species preferred for building materials (for standing poles, lattice structure, and roofing structure) are Terminalia sericea, Pteleopsis myrtifolia, Antidesma venosum, Croton pseudopulchellus, Brachylaena disco/or, Balanites maughamii, Hymenocardia ulmoides, Sapium integerrium, Monodora junodii and Uvaria lucida. 49

57 The need and harvesting of woody plants for fuel wood has probably the largest impact on the vegetation of the Licuati Forest Reserve. The Directorate of Forests and Wildlife of Mozambique has estimated that over 80% of the population of Maputo depends on fuel wood for their household cooking and heating needs. Most workers have indicated the high demand for this commodity as the major driving force behind deforestation and degradation of natural woodland in Mozambique. Saket ( 1994) estimated that deforestation was at 20 % around Maputo between 1970 and According to Chief Santaca, the major threat to the Licmlti Forest Reserve comes not from the local people, but rather from exploiters from the city of Maputo that are putting a lot of pressure on the plant resources of the reserve. There is an especially high demand for timber and charcoal production, and people from Maputo come to Licmiti to illegally cut timber and produce charcoal in high quantities to sell in Maputo. The proximity of the LFR to the economic development zone of the biggest city in Mozambique, Maputo, has put renewed pressure on the plant resources of the reserve for rapid economic development to carry the urban population. As a result of urbanisation, some of the people that resettle around the Licmiti Forest Reserve have no regard for cultural taboos, and hence the abundance and diversity of the large tree species have decrease remarkably the past few years (Izidine et al. 2002). Of the 113 vascular plant species recorded in the Licmiti Thicket, about 45 are used for medicinal purposes. This corresponds to about 40% of the total number of species in this vegetation type. The most common parts of plants collected are leaves, roots and bark. The more common medicinal plants are Balanites maughamii, Tarenna spp., Securidaca longipeduncu/ata, Synoptolepis kirki, Xylotheca kraussiana, Acridocarpus natalitiu var. /inearifolius, Erythroph/eum /asiathum, Brachy/aena huil/ensis, Ochna spp. and Sapium integerrimum. 50

58 According to Cunningham ( 1997), high levels of expectations, high rates of unemployment, and a psychologically stressful environment put increased pressure on the medicinal plant market, as many of the traditional medicinal plant and animal materials sold in urban market has a symbolic or psychosomatic value for luck in finding employment, guarding against jealousy, etc. Although the knowledge and use of medicinal plants are centred or dominated by traditional healers, it is not restricted to them. Every member of the community have at least a working knowledge of first aid treatments, particularly with regard to the most common diseases and medical conditions such as diarrhoea, fevers, and snake and insect bites. This knowledge is not the same as the more extensive knowledge of traditional healers that are passed on within the family (Mangue 1999). Traditional control and protection of the natural resources of the Licuati area Traditionally the people of the Licmiti live with many taboos, rites and believes associated with religion and the graves of their ancestors. According to Chief Santaca, the traditional ceremonies are very important, not only for his family, but for the whole Licmiti community. Taboos (swiyilayila) are regulators of resource use. Traditional harvesting practices, such as the harvesting of bulbous plants only after flowering and seeding have taken place, contributes to species conservation (Cunningham 1997). Certain taboos have also evolved prohibiting the use of some species. The use of Helichrysum krausii, which is burnt as incense, is, for example, controlled because of its religious value. There are also taboos on some fuel wood species such as Garcinia livingstonii, Tabernaemontana e/egans and Vangueria infausta. These species are pioneer and fruit bearing species. If a person disregards this taboo in the case of the death of the head of the family (m uti), it would result in the breaking up of the family of the person who broke the taboo (Mangue 1999). 51

59 The Santaca family, from which the current Acting Chief Augusto Santaca comes, is related to two historical chiefs that ruled in the area, namely Maputo and Tembe. Maputo and Tembe were two of the sons of the first King of Matuituine, Uangobe Tembe, who ruled in Kapizulu (also known as Uangobe). King Uangobe's third son, Panhela, took the surname Dlamini, and his descendants now live mostly in Swaziland. The traditional chiefs of the Santaca family play an important role in the control and protection of the natural resources of the Licmiti region. The family name Santaca originated when the second chief whose first name was Santaca, a descendent ofmaputo, took the Portuguese names of Jorge Oliveira as first names, and relegated his birth name, Santaca to surname status. The sacred place is a site in the LFR, where the Santaca family goes to pray and ask for rain, good harvesting and good luck for the whole community. This is an area covered in part by Licmiti Forest. This part of the forest is also the cemetery of the Santaca family and is off limits to other members of the community, unless they get special permission to enter and harvest there from the Chief. The Sacred Forest therefore plays an important role in the conservation of plant resources and of the rare and endemic species that occur there. As a consequence of the prohibition on entry into the Sacred Forest, a major part of the forest remains pristine (Mangue 1999). Trees in the Sacred Forest play a part in birth rites, initiation ceremonies and burial rituals. In stable rural communities, the cultural and spiritual roles of the vegetation, and the taboos, rites and believes associated with it, still persist, and are even reinforced in the face of encroaching values from outside the communities (Mangue 1999). Sclerocarea birrea subsp. caffra (marula) is the most highly valued tree in the culture of the Ronga ethnic group (to which the Santaca family belongs). As a result of this, when an area is cleared for agriculture timber or firewood Sclerocarya birrea subsp. caffra is often the only species left standing (Mangue 1999). The celebration of the Feast of the First Fruits of the marula symbolises the celebration of good crops. The elders of the family, the 'regulado', perform this feast by pouring a libation of the fresh juice of the marula fruit over the tombs of 52

60 their dead chiefs (Palgrave 1977; Mangue 1999). If the same kind of importance and protection that is afforded to the marula could be offered to other threatened species such as Afzelia quanzensis [25), Warburgia salutaris and Securidaca longipedunculata, it could have a major positive effect on the conservation of these species. CONCLUSIONS The name Licmiti Thicket is proposed for a distinct vegetation type in southern Mozambique. The largest, more or less continuous stands of Licuati Thicket, is contained the LFR. The vegetation of the Licuati Forest is characterised by tall, generally evergreen trees, while deciduous or semi-deciduous shrubs dominate the vegetation of the Licuati Thicket. Licuati Thicket contributes 51% of the total vascular plant diversity of the LFR, and harbous about 13% of the species endemic to the MC. Of the 113 indigenous species (of which more than 89% are dicotyledons) recorded for the Licuati Thicket, 30 species are endemic or near-endemic to the MC. Although at least 45 medicinal plant species of medicinal importances were recorded for the Licuati Thicket, most are usually collected outside of the thicket due to the very close and impenetrable characteristic of this vegetation type. This impenetrable nature of the thicket is one of the reasons why the Licuati Thicket is still in a relatively pristine state. Licmiti Thicket covers a significant area within the LFR, and it is clear that this vegetation type has remained stable for the past years with little evidence of 53

61 disturbance and virtually no indications of significant changes in the structure and extent of the vegetation. The structure and floristic diversity of Licmiti Forest, on the other hand, have changed more due to the illegal harvesting and over exploitation by outsiders mainly of the timber species Afze/ia quanzensis [25] (Chanfuta), Newtonia hildebrandtii and the fire wood and charcoal species Acacia nigrescens, Combretum imberbe, Albizia forbesii, Dichrostachys cinerea and Terminalia sericea. In the conservation of Licmiti ecosystem, taboos, rites and believes related to the religious and cultural life of the community, as well as the traditional leadership in the area as represented by the Santaca family, played an important role in the past, and can continue to do so in the future, through the impact of indigenous knowledge systems on management practices and resource use patterns in the area. The traditional leader of the area, Acting Chief Augusto Santaca, recognises that the unique vegetation and the endemic flora and fauna of LFR, and specially the Licmiti Thicket, are a valuable natural asset and could make the area an important ecotourism destination in southern Africa. 54

62 RECOMMENDATIONS The LFR should continue to play a role in the conservation of Afze/ia quanzensis-the main reason for which it was established in the first case. It is further recommended that the Licmiti Thicket, one of the reserve's ecosystems, should be afforded special protection due to its unique structure and floristic composition, diversity, high numbers of MC endemics and limited geographical range. Licuati Thicket should be recognised as a unique vegetation type endemic to the MC, which differs from Licuati Forest, and every effort should be made to ensure the preservation of this ecosystem for future generations. Existing legal roads that pass through the LFR should be kept open for law enforcement purposes, while the illegal roads should be closed down to lessen the pouching and illegal harvesting of the natural resources in the reserve. The local communities right to their natural resources must be recognised, thereby empowering them and encouraging them to maintain and use these resources on a sustainable basis. This must be one of the strategies to encourage the conservation of the natural resources of the Licuati area. In addition it is recommended that traditional knowledge systems (local communities) and law enforcement (government) should work together to ensure the conservation of natural resources. The role of traditional (local) leadership in the management of natural resources should be part of the strategy for the conservation of such resources in Mozambique. Licuati Thicket should be developed and promoted as a premier ecotourism destination in Mozambique. It is not only rich in rare plant species, but also harbours many uncommon animals, notably birds. 55

63 Further research should be done in the area to gather more information that can be used to make people aware of the uniqueness of this botanical treasure in Mozambique, and to aid in the development of vegetation dynamics models to guide reserve management, land-use planning, and sustainable utilisation programmes. These tools will facilitate the conservation of these biologically rich, endemic vegetation types for future generations by taking into account both the socio-economic development and the conservation needs of the region. ACKNOWLEDGEMENTS The financial and material support from the Southern African Botanical Diversity Network (SABONET) and the University of Pretoria, as well as the provision of research facilities by these institutions are gratefully acknowledged. Dr Stefan Siebert, SABONET Regional Coordinator, was especially supportive and his help is acknowledged with gratitude. Special thanks also to Prof. Braam van Wyk for all orientation and support given during my studies, the Director of INIA, Dr Calisto Bias, and the Mozambique SABONET Coordinator, Mr Calane da Silva, for all support given during my research in Mozambique. I would also like to extend my gratitude toward Chief Augusto and Mr Edmundo Santaca, and the traditional healer Mr Dinis Nguenha, for giving their time for interviews and for helping me with my fieldwork. Many thanks also to Mrs El sa van Wyk, Ms Magda N el of the PRU Herbarium, Elizabeth Retief and Tandiwe Nkonki from PRE Herbarium in Pretoria, the late Mr. Salomao Macitela, Mr. Abilio Manhique, Mr. Alfredo Vilanculos from LMA Herbarium, Mr.Carlos Boane from LMU Herbarium in Maputo who assisted me during my herbarium studies and field work respectively and to Dr Adelia Diniz from LISC Herbarium, who sent me the scan of the holotype of Xylopia torrei, housed at LISC. Thank you also to Y olande Steenkamp who improved the English of this report. 56

64 REFERENCES CUNNING HAM, A.B An Africa-wide overview of medicinal plant harvesting, conservation and health care. In: FAO ( ed) Medicinal plants for forest conservation and health care, Non-Wood Forest Products. N.ll.Roma. EDWARDS, D A broad-scale structural classification of vegetation for practical purposes. Bothalia 14: IZIDINE, S., SIEBERT, S. & VAN WYK, A.E Remarkable vegetation types in southern Mozambique, paper presented at SAAB conference. Unpublished. MANGUE, P Community use and management oflicuati Reserve and surrounding areas. Paper presented at the Workshop on Integrated Analysis and Management of Renewable Natural Resources in Mozambique, Maputo, 7-11 June 1999, (eds) P.V. Desanker & L.Santos. Project/Mangue.pdf (Cosulted in September 2002) SAKET, M Report on the updating of the exploratory national forest inventory. FAO/UNDP/MOZ/92/0113, Maputo. SIEBERT, S.J., BANDEIRA, S.O., BURROWS, J.E. & WINTER, P.J SABONET southern Mozambique expedition SABONET News (March) 7(1): V AN WYK, A.E. & SMITH, G.F Regions of floristic endemism in southern Africa. A review with emphasis on succulents. Umdaus Press, Pretoria. VAN WYK, A.E Biodiversity of the Maputaland Centre. In: (eds.) L.J.G. van der Maesen, X.M. van der Burgt, & J.M. van Medebach de Rooy. The Biodiversity of African Plants. Kluwer Academic Publishers, Dordrecht. 57

65 ANNONACEAE THE IDENTITY OF XYLOPIA TORREI INTRODUCTION The genus Xylopia L. belongs to the Annonaceae, a family, which consists of a shrubs and trees characterised by simple, alternate leaves, with stellate or branched hairs or trichomes. Members of the Annonaceae bear single, indehiscent or dehiscent fruit that may, in some cases, appear to be a cluster of several fruits, but which are, in fact, the separate carpels or segments of the same ovary, each one developing independently from the others (Palgrave 2002). The Annonaceae consists of about 100 genera, with Xylopia the only one that is pantropical. Of its c. 160 tropical species, about 61 occur in Africa, and 51 in America. Currently eight species of Xylopia are recognised in southern Africa, most of them in the Flora zambesiaca (FZ) region. They are: Xylopia rubescens Oliv., X aethiopica (Dunal) A.Rich, X acutiflora (Dunal) A.Rich., X katangensis De Wild., X torrei N.Robson, X odoratissima Welw. ex Oliv., X tomentosa Ex ell, and X collina Diels. Habit, leaf shape, hairiness of the leaves, lamina size, length of the pedicels, length of the petals, fruit characteristics, and the arrangement of the seed in the fruit are the main characters that are used to distinguish among the nine southern African species of Xylopia. Despite the use of these characters in the available identification keys, herbarium workers often find it difficult to identify the different species. The difficulty to identify species is aggravated by the many misidentified specimens of Xylopia in herbaria. Four species of Xylopia have been reported in Mozambique. These are X aethiopica, X parviflora, X torrei [12,13,14] and X collina. Until recently, X torrei was known from only three collections by Torre made at Chibuto and Panda in Gaza, and Inhambane provinces respectively. It has first been described as a new species by Robson (1958), and was 58

66 subsequently taken up in the species list of Lebrun & Stork ( 1991 ). Xylopia torrei, was considered by Robson ( 1960) to be closely related to X holtzii Engl., which according to Pal grave (2002), is a synonym of X parviflora. Habit, branch colour, the size of the lamina, the hairiness of the leaves, and the type of seed distinguishes it from X parviflora (=X holtzii). Differences between X torrei and X parviflora, as well as a comparison of these two species with X odoratissima, are listed in Table 1. Literature studies and observed characteristics, however, suggest that X torrei is more closely related to X odoratissima than to Xparviflorai (Table 4). Very few specimens of X torrei [12, 13, 14] have ever been collected, and the identity of this species is unclear. As it has been described as a shrub, and has not been taken up in the standard book of Mozambican trees (Games e Sousa 1960). Because so little is known about the species, it has provisionally been listed in the Red Data List for Mozambique (Izidine & Bandeira 2002), and is catagorised as Lower Risk-least concern (LR-lc). During a Southern African Botanical Diversity Network (SABONET) expedition to southern Mozambique from 22 November 2001 to 12 December 2001, an unexpected discovery of a Xylopia species was made whilst collecting in the Licmiti Forest Reserve. This was the first record of Xylopia in southern Mozambique. The closest previous records of Xylopia torrei are from Chibuto, A.R. Torre 3944 in 1942 and Xylopia parviflora from the Kruguer National Park (Van Wyk 1984). 59

67 Table 4. Comparisson between X torrei, X parviflora, and X odoratissima. Characters X torrei X parviflora (=Xholtzii) X odoratissima Habit Shrub up to 2 m high Tall tree up to 30 m high Shrub or small tree (1-) 2-6 (-9) m high. Branch Red-brown, turning Red to purplish-black, Red-brown, densely colour greyish, eventually sparsely yellowish-fawn- yellowish-fawn-pubescent glabrous pubescent at first, soon at first, eventually glabrous. glabrous. Leaf shape Elliptic or oblong to Narrowly oblong to elliptic, Elliptic to oblong, rarely suborbicular, obtuse or acute or more rarely obtuse or lanceolate, obtuse or rarely rounded rounded subacute to rounded. Hairiness of Glabrous or sparsely Sparsely appressedleaves appressed-pubescent above. pubescent, rarely glabrous above. Lamina size x mm x mm (-135) x mm Pedicel 4-5 mm long 4-8 mm long Absent or up to 7 mm long length Length of the Outer: mm long Outer (- 25) mm long Outer mm. long petals Fruit On pedicel c. 8 mm long, On a pedicel c. 5-8 mm long, On a pedicel 4-7 mm long, characteristic carpels. carpels x 7-11 (- 15) carpels x 8-11 mm. mm long. Type of 5-seeded, subglobular, 1-6 seeded, obovoid or 1-5 seeded, obovoid seeds with stipe 1-2 mm.long. cylindric, with stipe 2-9 mm or cylindric, with stipe long, seeds 8-11 mm long, 3-4 mm long, ellipsoid, biseriate. seed 12 mm long, ellipsoid, biseriate. 60

68 The fruit of the Xylopia described by Rob son ( 1958) as X torrei were immature and contained one seed per fruit. The fruit of the Xylopia that were collected in the Licuati Forest Reserve during the SABONET expedition were mature and contained several seeds. The purpose of this paper is to clarify the identity of Xylopia torrei [12, 13, 14], and to provide an amplified description of the species, using evidence from macromorphology, leaf anatomy, and leaf surface structure (SEM studies). MATERIAL AND METHODS Material of Xylopia from the following herbaria was studied: LISC, LMA, PRE, and PRU. For the study of the leaf anatomy and leaf surface structure, both fresh and rehydrated leaves were used. Leaf anatomy was studied using light microscopy (LM), and leaf surface structure by using a JEOL 640 scanning electron microscope (SEM). Xylopia torrei N.Robson in Bol. Soc. Brot., Ser. 2, 32: 157 (1958). Type: Mozambique, Gaza, Chibuto, Torre 2350 (LISC, holotype). Shrub up to 2 m high, branches reddish-brown, turning greyish, eventually glabrous. Leaves petiolate; lamina x mm, oblong to elliptic to suborbicular, obtuse or rounded to shallow emarginated at the apex, broadly cuneate to rounded at the base; petiole 2 mm long. Flowers solitary or more rarely paired, spreading or reflexed; pedicels 4-5 mm long. Outer petals mm long. Capels 4-7 mm. Fruits on pedicel c. 8 mm long; fruiting carpels c x 5 mm, 5-seeded, subgobular, with stipe 1-2 mm long. 61

69 CONCLUSIONS All material examined in terms of both macromorphology and anatomy show that Xylopia tomentosa (fig. 9 & 1 0) is a unambiguous, clear species. It is still difficult to distinguish between the other species, especially between X parviflora (=X holtzii) and X odoratissima (7 & 8 and 5 & 6). Anatomical differences were found between X torrei and X tomentosa. Only X tomentosa differs significantly from the other species. No significant anatomical differences were found between the other five species studied. The species found in the Licmiti Forest Reserve is provisionally recognised as X torrei according to the type specimen. The slight leaf anatomical differences found between the leaves of X torrei (12, 13, 14) collected in the Licmiti Forest Reserve and the other species may be due to the different ages of the leaf samples, and the fact that some of the rehydrated leaves samples were from old herbarium specimens. ACKNOWLEDGEMENTS I would like to express my sincere gratitude to my supervisor Prof. Braam van Wyk for his guidance. Many thanks also to Mrs Eisa van Wyk and Magda Nel for help in the H.G.W.J. Schweikerdt Herbarium, to Mrs Elizabeth Retief and Tandiwe Nkonki who assist me at National Herbarium and Mr Chris van der Merwe and Mr Allan Hall who help me with SEM work. 62

70 SPECIMENS EXAMINED Xylopia torrei N. Rob son Van Wyk 4332 (PRU) Van Wyk 4345 (PRU) A.R. Torre 3944 (LMA) A. R Torre 2350 (LISC) Xylopia odoratissima Welw.ex Oliv. Stalman, S.M. 285 (PRE) D.C.Mc.Ferren 30 (PRE) F. V. Breitenbach 1312 (PRE) S. Kandanda 11 (PRE) Harder, MBingham & Luwiika 829 (PRE) Xylopia parviflora (A. Rich.) Benth H.P.v.d Schyf/3779 (PRE) F. V. Breitenbach (for Herb) Bremekamp & Schweickerct 348 (PRE) A.R.Torre, & MF. Correia (LMA) J. Pedro Games 4224 (LMA) M.F. Carvalho 685 (LMA) Goldsmith, B. 159/68 (PRE) Xylopia tomentosa Exell B. Winter 3845 (PRE) B. Winter 3879 (PRE) W Giess (PRE 63

71 Figures SEM micrographs of leaf lower surfaces. 3 & 4Xylopia torrei (Van Wyk 4332). 5 & 6 X odoratissima (D.C. Me. Ferren 30). 64

72 Figures SEM micrographs of leaf lower surfaces. 7 & 8 Xylopia parviflora (M. F. Carvalho 685). 9 & 10 X tomentosa (B. de Winter 3845). 65

73 REFERENCES FERNANDES, A. & NEVES, B.J Boletim da Sociedade Broteriana. Instituto Botanico da Universidade de Coimbra Vol XXXII (2.A SERIE). Coimbra, Portugal. IZIDINE, S. & BANDEIRA, S.O Mozambique. In:J.S.GOLDING (ed.). Southern African Plant Red Data Lists. Southern African Botanical Diversity Network Report N 14: 54. SABONET. Pretoria. LEBRUN,P.J. & STORK, L.A Enumeration Des Plantes A Fleurs D'Afrique Tropicale, I Generalites et Annonaceae a Euphorbiaceae et Pandaceae. Ville de Geneve, Editions des Conservatoire et Jardin botaniques, Geneve 40. PALGRA VE, C. M Trees of Southern Africa. Struik Publishers, Cape Town. ROBSON, N Anonnaceae. Flora Zambesiaca, Vol. (1): SOUSA, A.G Dendrologia de Movambique. Maputo, Mozambique. VAN WYK, P Trees ofkruger National Park. Struik Publishers, Cape Town. 66

74 SUMMARY Licu~lti Forest Resrve: flora, vegetation and utilisation by Samira Aly Izidine Supervisor: Prof. Dr. A.E. van Wyk Department of Botany University of Pretoria Magister Scientiae (Systematics and Conservation Evaluation) The Licmiti Forest (usually referred to as Sand Forest in South Africa) is a very distinctive vegetation type in southern Mozambique. This dry tropical type of forest only occurs in southern Mozambique and northern KwaZulu-Natal, South Africa. The unique combination of rare plants, endemic species and its restricted global distribution make this vegetation type one of the most important plant communities of the Maputaland Centre of Plant Endemism. The short dense form oflicuati Forest also known as short Sand Forest covers about ha in the Licuati Forest Reserve in southern Mozambique. It displays a unique vegetation structure, floristic composition and high levels of endemism. Based on Edwards' structural classification of vegetation, we proposed the name Licmiti Thicket for this vegetation type. The proximity of this Licuati Forest and Thicket in southern Mozambique to the city of Maputo is putting considerable pressure on the plant resources of the region. Due to the impenetrable nature of Licuati Thicket, this vegetation type has remained stable over the past 30 years. The establishment of the Licmiti Forest Reserve was 50 years ago on 14 December 1943 by portaria number 5354, to protect woody species of economic value, especially Afzelia quanzensis [25] (Chanfuta). The principal aim of this study was to provide floristic, ecological and etnobotanical information on the Licmiti Thicket. 67

75 The flora of the Licmiti Thicket comprises about 113 indigenous species, most (89%) being dicotyledons. Monocots and ferns make up 4.5o/o and 15% respectively of the total flora of the Thicket. The families of flowering plants are very well represented with Rubiaceae (coffee family) contributing the most species and genera; it is also the most prolific family in terms of endemic species. The Licmiti Thicket contains ea. 30 endemic species, or 13% of the total number of species/infraspecific taxa endemic/near-endemic to the Maputaland Centre. The Licuati Thicket is characterised by shrubs such as Acridocarpus natalitius subsp. linearifolius, Artabotrys monteiroi, Monodora junodii, Uvaria lucida, Dialium schlechteri [18], Hymenocardia ulmoides and Warneckea sousae which contrast with the species composition of the Licuati Forest which is subject to more utilisation and where tall trees such as Afzelia quanzensis [25], Newtonia hildebrandtii, Pseudobersama mossambicensis, and Albizia adianthifolia occur. Medicinal plants correspond to about 40% of the total number of species in the Licuati Thicket. The more common medicinal plants are Balanites maughamii, Tarenna spp., Securidaca longipedunculata, Synaptolepsis kirki, Xylotheca kraussiana, Acridocarpus natalitius var.linearifolius, Erythrophleum lasiathum, Brachylaena huillensis, Ochna spp. and Sapium integerrimum. A checklist is given of the 113 plant taxa recorded in the Licuati Thicket (Appendix 1), arranged by family, with genus and species listed alphabeticaly within, with attributes such as common name, growth form, life history, fruit type and medicinal uses. The identy of Xylopia torrei [12, 13, 14] is confirmed. It is the first record of a Xylopia for southern Mozambique. Leaf anatomy and macromorphology suggest that the species is closely related to Xylopia parviflora and Xodoratissima. Based on historical information provided by Chief Santaca and aerial photographs from 1958, 1982 and 1989, the vegetation structure of the Licuati Thicket has not changed in living memory. Traditionally the people in the Licuati Thicket region lived with many taboos, rites and believes associated with religion and the graves of their ancestors. Certain taboos have also evolved prohibiting the use of some species and could play an important role in the conservation of the Licuati ecosystem. Available data on vegetation structure, floristic composition, and levels of endemism supports the recognition of the Licuati Thicket as one of the most important plant communities of the MC. All possible efforts should be made to conserve this biologically rich vegetation type for future generations. Efforts towards this end should take into account both the socio-economic development and conservation needs of the region. 68

76 GENERAL ACKNOWLEDGEMENTS Thanks are due to the following institutions and persons for their valuable contribution in the preparation of this report: Prof. Braam van Wyk for his invaluable guidance. Dr Stefan Siebert, SABONET, National Botanical Institute, Pretoria, for all his support and encouragement; Dr Calisto Bias, National Institute of Agriculture Research (INIA), Maputo, for the support given during my research in Mozambique; Mr Mario A. Calane da Silva, National Institute of Agriculture Research (INIA), Maputo, for the support given to make possible my field and herbarium work in Mozambique; My husband Julio Oliveira, our children Yuri, Marua and little Shakila for all their unconditional love, encouragement, support and patience during my long absence from Mozambique; Dr Adelia Diniz, LISC Herbarium, Lisbon, for her valuable collaboration and for sending me the scan of the holotype of Xylopia torrei housed at LISC; Mrs Eisa van Wyk, Department of Botany, University of Pretoria, for handling the administration of this project; Ms Magda Nel, Department of Botany, University of Pretoria, for technical assistance in handling the collected plant speciemens; Ms Elizabeth Retief, National Botanical Institute, Pretoria, for scientific and technical support; Mr Alan Hall, University of Pretoria, for assistance in the use of Scanning Electron Microscopy [SEM]; Mr Chris van der Merwe, University of Pretoria, for assistance with the SEM work; Mrs Tandiwe Nkonki, National Botanical Institute, Pretoria, for specimens identification support; Mrs Nyasha Leboho, SABONET, National Botanical Institute, Pretoria, for solving all administrative issues regarding to enrolment for my studies; 69

77 Mrs Loma Davis, for solving all the financial issues during my stay in Pretoria; Mrs Regina Cruz, Inventory Forest Centre, Maputo, for valuable discussions on the Licmiti Forest Reserve; Mr Poncio Maciel, Department of Land and Water, National Instituto of Agricultural Reserch (INIA), Maputo, for assisting me with the interpretation of aerial photographs; Mr Castelo Jonas Banze, Inventory Forest Centre, Maputo, who provide the Vegetation and Topographic Maps of the LFR; Mr Moniz Cumbane, National Directorate of Agriculture, Maputo, who supplied a copy of the government gazette in which the LFR was proclaimed; Chief Augusto Santaca, Tinonganine, for being very supportive and for giving his time for interviews, Mr Edmundo Santaca, Tinonganine, for his collaboration during our visit to the Santaca sacred site and for the interview; Mr Dinis Nguenha, Tinonganine, for his help on the identification of plants with medicinal purposes and for the interview; The late Mr Salomao Macitela, Department of Botany, National Institute of Agricultural Reserch (INIA), Maputo, for helping me during fieldwork; Mr Abilio Manhique, Department of Botany, National Institute of Agricultural Reserch (INIA), Maputo, for handling the specimens during the fieldwork; Mr Alfredo Vilanculos, Department of Botany, National Institute of Agricultural Reserch (INIA ), Maputo, for driving me during fieldwork; Mr Carlos Boane, Department of Botany, Eduardo Mondlane University, for assisting me during herbarium studies at LMA Herbarium, Maputo; Ms Y olande Steenkamp, SABONET, National Botanical Institute, Pretoria, for improvements to the English of the present Report; Ms Frances Siebert, University of Pretoria, for the help given during the compilation of the report; Mr Willem de Frey, University of Pretoria, for assistance with the preparation of the map of the Licmiti Thicket and surrounding areas; 70

78 Southern African Botanical Diversity Network (SABONET), for financial support; The National Directorate of Geography and Cadastre, Maputo, for supplying the aerial photographs; The National Centre of Cartography and Teledating, Maputo, for providing the Landsat images. 71

79 CURRICULUM VITAE Samira Aly Izidine was born in Inhambane city, Mozambique, on 20 January She attended 7 April Primary School and later Emilia Dausse High School in Inhambane, where she graduated in In 1983 she attended the Francisco Manyanga Pre-University School in Maputo where she finished the pre-university studies (Matric) in In 1986 she enrolled at the Eduardo Mondlane University, Maputo, and was awarded her B.Sc. Honours degree in 1994, with Botany and Ecology as majors. Samira registered for a Master's degree in Botany at the University of Pretoia in 2002, specialising in Systematics and Conservation Evaluation. In 1992 she joined the LMA Herbarium, INIA, Maputo as Research Officer, a position she still holds. In 1998 she was appointed Plant Red Data List Mozambique Coordinator as part of a SABONET Project. As Red Data List Coordinator she established the National Red Data List Working Group and organised the first workshop on the dissimination and implementation of Red Data List Concepts in Mozambique. Presently Samira is involved in the compilation of a Checklist of the Plants of Mozambique. She is also preparing the taxonomic treatments of Comaceae and Begoniaceae for the Flora of Mozambique. Apart from plant taxonomy, her interests also include etnobotany, plant ecology and nature conservation. 72

80 GENERAL REFERENCES BOTHA, G.A The Maputaland group: a provisionallithostratigraphy for coastal KwaZulu-Natal. In: Botha G.A. (ed.) Maputaland: Focus on the Quaternary Evolution of the South-East African Coastal Plain, field guide and abstracts. Council for Geoscience, Pretoria, South Africa, BRUTON MN, COOPER KH (eds) (1980) Studies on the Ecology ofmaputaland. Rhodes University and Wildlife Society of South Africa, Grahamstown and Durban, South Africa. CRAIB, A Maputaland. Africa-Environment & Wildlife 3 (4): CUNNINGHAM, A.B An Africa-wide overview of medicinal plant harvesting, conservation and health care. In: FAO ( ed.) Medicinal plants for forest conservation and health care, Non-Wood Forest Products 1l.Roma, DAVIS, S.D., HEYWOOD, V.H. & HAMILTON, A.C Centres of Plant Diversity: A guide and strategy for their conservation. IUCN Publications Unit, Cambridge. DE MOOR, P.P., POOLEY, E., NEVILLE, G. & BRARICHIEVY, J The vegetation ofnduma Game Reserve, Natal: a quantitative physiognomic survey. Annuals of the Natal Museum 23: DUTTON, P. & DUTTON, S Looking beyond the healing values of medicinal plants in Mozambique. HYPERLINK: ' I 121 /wild2nov.html' (Consulted in March 2003). EDW ARDS, D A broad-scale structural classification of vegetation for practical purposes. Bothalia 14:

81 FERNANDES, A. & NEVES, B.J Boletim da Sociedade Broteriana. Instituto Botanico da Universidade de Coimbra Vol XXXII (2A SERIE). Coimbra, Portugal. SOUSA, A.G Dendrologia de Moc;ambique. Maputo, Mozambique. GOODMAN, P.S Soil, vegetation and large herbivore relations in the Mkuzi Game Reserve, Natal. PhD thesis, University of Natal, Pietermaritzburg, South Africa. HOPKINS, B Forest and savanna. An introduction to tropical terrestrial ecology with special reference to West Africa. 2nd edition. Heinemann, London. IZIDINE, S., SIEBERT, S., & V AN WYK, A.E Remarkable vegetation types in southern Mozambique, paper presented at SAAB conference, Pretoria. Unpublished. IZIDINE, S. & BANDEIRA,S.O Mozambique. In: J.S.GOLDING (ed.). Southern African Plant Red Data Lists. Southern African Botanical Network Report N 14: 54. SABONET, Pretoria. KIRKWOOD, D. & MIDGLEY, J.J The floristics of Sand Forest in northern KwaZulu-Natal, South Africa. Bothalia 29: LEBRUN, P.J. & STORK, L.A Enumeration Des Plantes A Fleurs D' Afrique Tropical, I Generalities. Annonaceae a Euphorbiaceae et Pandaceae. Ville de Geneve, Editions des Conservatoire et Jardin botaniques, Geneve 40. LOW, A.B. & REBELO, A.G. (eds.) Vegetation of South Africa, Lesotho and Swaziland. Dept. Environmental Affairs & Tourism, Pretoria, South Africa. 74

82 LUBBE, R.A Vegetation and flora of the Kosi Bay Coastal Forest Reserve in Maputaland, northern Kwazulu-Natal, South Africa. M.Sc. thesis, University of Pretoria, Pretoria. MANGUE, P Community use and management oflicmiti Reserve and surrounding areas. Paper presented at the Workshop on Integrated Analysis and Management of Renewable Natural Resources in Mozambique, Maputo, 7-11 June (eds) P.V. Desanker & L.Santos. Project/Mangue.pdf (Consulted September 2002). MARKER, M.E. & RUSSEL, S The application ofbiogeographic techniques to forest site-factor analysis. South African GeographyJourna/66: MAUD, R.R., BOTHA, G.A Deposits of the south eastern and southern coasts. In: Partridge, T.C., Maud, R.R. (eds.), The Cenozoic of Southern Africa. Oxford University, Press, Oxford, UK, MATTHEWS, W.S., VAN WYK, A.E. & V AN ROOYEN, N Vegetation of the Sileza Nature Reserve and neighbouring areas, South Africa, and its importance in conserving the woody grasslands of the Maputaland Centre of Endemism. Bothalia 29: MATTHEWS, W.S., VAN WYK, A.E., VAN ROOYEN, N. & BOTHA, G.A Vegetation of the Tembe Elephant Park, Maputaland, South Africa. South African Journal of Botany 67: MIDGLEY, J.J., COWLING, R.M., SEYDACK, A.H.W. & VAN WYK, G.F Forests. In: R.M. Cowling, D.M. Richardson & S.M. Pierce (eds.). Vegetation of Southern Africa. Cambridge University Press, Cambridge,

83 MIDGLEY, J.J Biogeographic and other perspectives on the management and conservation of forests patches. In: Edwards D ( ed.) The Relevance of Island Biogeographic Theory in Commercial Forestry. Foundation for Research Development, Pretoria, South Africa, MOLL, E.J. & WHITE, F The Indian Coastal Belt. In: M.J.A. Werger (ed.). Biogeography and Ecology of Southern Africa. Junk Publishers, The Hague, MOLL, E.J. 1978a. A quantitative floristic comparison of four Natal forests. South African Forestry Journal! 04: MOLL, E.J. 1978b. The vegetation ofmaputaland. A preliminary report of the plant communities and their present and future conservation status. Trees in South Africa 29: MOLL, E.J Terrestrial Plant Ecology. In: M.N. Bruton & K.H. Cooper ( eds.). Studies on the ecology of Maputaland. Rhodes University & Wildlife Society of South Africa, Grahamstown & Durban, South Africa, PALGRA VE, C.M Trees of Southern Africa. Struik Publishers, Cape Town. POOLEY, E The complete field guide to trees ofnatal, Zululand and Transkei. Natal Flora Publications Trust, Durban. ROBSON, N Anonnaceae. Flora Zambesiaca, Vo1.(1): SAKET, M Report on updating of the exploratory national forest inventory. FAO/UNDP/MOZ/92/0113, Maputo. 76

84 SCHULTZE, R.E Agrohydrology and agroclimatology of Natal. Agricultural catchments Research Unit, Report N 14. University of Natal, Pietermaritzburg. SOUSA, A.G Dendrologia de Mo9ambique. Maputo, Mozambique. SOUSA, A.G Reserva Florestal do Licuati. Instituto de Investiga9ao Agron6mica de Mo9ambique, Comunica96es 13, SIEBERT, S.J., BANDEIRA, S.O., BURROWS, J.E. & WINTER, P.J SABONET southern Mozambique expedition SABONET News 7 (1): TINLEY, K.L The moist evergreen forest-tropical dry semi-deciduous forest tension zone in the north-eastern Zululand and hypotheses on past temperate/montane rain forest connections. In: (ed.) E.M. van Zinderen Bakker. Unpublished report, VAHRMEIER, J Notes on the vegetation of northern Zululand. African Wildlife 20: V AN ROOYEN, N., THERON, G.K., GROBBELAAR, N A floristic description and structural analysis of the plant communities of the Punda Milia-Pafuri-Wambiya area in the Kruger National Park, Republic of South Africa: 2. The sand veld communities. Journal of South African Botany 47: VAN WYK, A.E Maputaland-Pondoland Region. In: Centres of plant diversity, (eds) S.D.Davis, V.H Heywood & A.C Hamilton.Volume 1. njcn Publications Unit, Cambridge, VAN WYK, A.E Biodiversity of the Maputaland Centre. In: (eds.) L.J.G. van der Maesen, X.M. van der Burgt, & J.M. van Medebach de Rooy. The Biodiversity of African Plants. Kluwer Academic Publishers, Dordrecht. 77

85 VAN WYK, A.E. & SMITH, G.F Regions offloristic endemism in southern Africa. A review with emphasis on succulents. Umdaus Press, Pretoria. V AN WYK, P Trees ofkruguer National Park. Struik Publishers, Cape Town. WARD, M. C A preliminary investigation into the ecology of Sand Forest, Maputaland. B.Sc. (Hons.) dissertation, University of Natal, Durban. WHITE, F The afromontane region. In: (ed.) M.J.A. Werger. Biogeography and ecology of southern Africa. Junk Publishers, The Hague, WHITE, F The vegetation map of Africa. A descriptive memoir to accompany the Unesco/AETFATIUNSO Vegetation Map of Africa. UNESCO, Paris, France. 78

86 APPENDIX 1 ANNOTATED CHECKLIST OF PLANTS OF LICUATI THICKET, MOZAMBIQUE 79

87 APPENDIX 1. ANNOTATED CHECKLIST OF PLANTS OF LICUATI THICKET, MOZAMBIQUE For an explanation of attributes and abbreviations, see end of list PTEROPHYTA PTERIDACEAE Cheilanthes viridis (Forssk.) Swartz var. viridis, SI 224 PRU liseka-seka(r) Eg H;fem Water of boiled leaves is taken three times daily to treat venereal diseases. SELAGINELLACEAE Selaginella sp., SI 225 PRU Eg H;fem LYCOPODOPHYTA AMARANTHACEAE Cyathula natal ens is Sond SI 218 PRU Eg H;D ANGIOSPERMAE-MONOCOTYLEDONAE ARACEAE Zamioculcas zamiifolia (Lood.) Engl., SI 181 PRU fern arum(e) De H; F(berry) ASPARAGACEAE Asparagus africanus Lam., SI 203 PRU bush asparagus(e); isigobo(z) Eg Shlt; F(berry) Used to treat kidney and stomach complains, chest infections, nausea, colic and as a protective charm. 80

88 densiflorus (Kunth.) Jessop, SI 231 PRU emerald fem(e); uvelabahleke(z) Eg Sh; F(berry) setaceus (Kunth.) Oberm, SI 136 PRU cangulatilo(r); ibutha(z) Eg Sh; F(berry) Used against human lice. The pounded roots are boiled. Baths are taken two times daily. ASPHODELACEAE Aloe sp., SI 214 PRU chiramacha (R) Eg; D( capsule) Used to treat venereal diseases. Pieces of the root are boiled and decoctions taken three times daily. CYPERACEAE Cyperus albostriatus Schrad., SI 202 PRU forest star sedge(e); selepusamboti (R); ephakathi(z) Eg H; D(nut) DRACAENACEAE Sansevieria metallica Gerome & Labroy, SI 158 PRU mother-in-law's-tongue(e); isitokotoko(z) Eg Sh; F(berry) Used as remedy for earache, toothache, stomach pains and sexual impotence. Also used to treat intestinal worms, haemorrhoids and as protective charm. Decoction or infusions of the roots and leaves are taken three times daily. POACEAE Eragrostis moggii var. moggi De Winter, SI 208 PRU (e) Pe G; D( caryopsis) Panicum deustum Thunb., SI 222 PRU Broad-leaved panicum(e); ipunga(r) Pe G; D( caryopsis) 81

89 ANGIOSPERMAE-DICOTYLEDONAE ANNONACEAE Artabotrys monteiroae 0/iv., SI 164 PRU red hook-berry(e); okonjowati(r); umazwenda(z) De St/sh/1; F(berry) Monodora junodii Engl & Diels, SI 165 PRU green apple(e); coni(r); umkotshi(z) De St/sh, F(berry) Uvaria caffra E. Mey. ex Sond., SI 205 PRU small cluster-pear(e); umazwenda-omnyama(z) Eg St/sh/cl; F (berry) lucida Benth., SI 217 PRU large-cluster-pear(e); umavumba(z) Eg St/shlc; F (berry) Used to relieve general pains in the body. Roots are boiled and the decoction is taken three times daily. Xylopia torrei N.Robson, SI 197 PRU [12, 13,14] Xylopia(E) Eg St/sh; F(berry) Used to treat rheumatism. The decoction is taken three times daily. APOCYNACEAE Carissa tetramera (Sacleux.) Stapf, SI 152 PRU sand carissa(e); chivelawyelane(r) Eg Sh/st; F(berry) Landolphia kirkii Dyer, SI 154 PRU bungue(r); umbungwa(z) De shlcl; F(berry) Used as anthelmintic. Roots are boiled in water and the infusion is taken orally three times daily as a tea. 82

90 Strophanthus gerrardi Stapf, SI 219 PRU (e) poison-rope(e); umhlazazane(z) Eg St/sh; D(follicle) luteolus Codd., SI 192 PRU golden poison-rope(e) Eg Mt; D(follicle) Used as remedy for general body pains. The decoction is taken three times daily. petersianus Klotzsch., SI 220 PRU corky-stem poison-rope(e); ubuhlungubendlovu(z) Eg Lt; D(follicle) Used as a charm against evil. Baths are taken two times daily. Taberrnamontana elegans Stapf, SI 182 PRU toad-tree(e); caxuana(r); umkhahlwana(z) Dc/SDc St/sh; F(berry) ASCLEPIADACEAE Sarcostemma viminale (L.) R. Br., SI 186 PRU Caustic vine(e); lineata (R); umpelepele(z) De H; D(follicle) Used to treat venereal diseases, as diuretic and to increase milk during lactation. Water of boiled roots is taken three times daily. Secamone delagoensis Schltr., SI 206 PRU (e) ikhoka (R) Eg Cl; D(follicle) 83

91 ASTERACEAE Brachylaena huillensis 0. Hoffm., SI 215 PRU lowveld silver oak(e); iphahlalahlathi(z) Eg St/mt; D(nut) Helichrysum krausiii Sch. Bip., SI 213 PRU chirimbati(r) Eg Sh, D(nut) Used as ritual incense, called imphepho. The smoke of roots and leaves are used in small babies to avoid epilepsy. Senecio barbertonicus Klatt., SI 207 PRU succulent bush senecio, jelly-beans (E); Eg Sh; D(nut) viminalis Bremekamp, SI 223 PRU leafless climbing senecio(e); uqobaqoba(z) Eg Cl; D(nut) BALANITACEAE Balanites maughamii Sprague, SI 159 PRU torchwood(e); chicocuane(r); ugobendlovu(z) De Lt/mt; F( drupe) Used to strengthen the body and as ritual emetic. Bark is applied in the form of sub-cutaneous implantations to strengthen the body and the root bark is mixed with other ingredients for use as emetics. CAPPARACEAE Boscia foetida Schinz subsp. longipedicellata (Gilg.) To/ken, SI 13 7 PRU smelly shepherd tree(e); umvithi(z) Eg St/sh; F (berry) foetida Schinz subsp. rehmanniana (Pest.) To/ken, SI 198 PRU stink shepherd's tree(e); umlalampisi(z) Eg St; F(berry) 84

92 salicifolia 0/iv., SI 226 PRU willow-leaved shepherds-tree( ), jose(z) De St/mt; F(berry) sp. nov., SI 169 PRU Capparis sepiaria L. SI 190 PRU wild caper-bush(e); usondeza(z) Eg St/shlcl; F(berry) CELASTERACEAE Hippocratea delagoensisloes., SI 173 PRU (e) shicocumbela (R) St/shlcl; D(pod) Used to treat rheumatism. The decoction is taken three times daily. Maytenus undata (Thunb.) Blakelock, SI 175 PRU dune koko tree(e); mugumo(r); idohame(z) Eg Lt/sh; D(capsule) Salacia krausii (Harv.) Harv., SI 227 PRU (e) salacia(e); mabsisa(r); mbontsi(z) Eg St/sh; F(berry) leptoclada Tu!., SI 170 PRU lemon rope(e); uhlangahomo(z) Eg St/sh; F(berry) COMBRETACEAE Combretum celastroides Welw. ex MA. Lawson, SI 143 PRU savanna bushwillow(e); mbabolile(r) SDc St/sh; D(samara) mkuzensej.d.carr & Relief, SI 168 PRU (e) (15] bushwillow(e); mkuze(z) De St/t; D(samara) 85

93 Pteleopsis myrtifolia (M.A. Lawson) Engl. & Diels, SI 133 PRU stink bushwillow (E); amuyadi (R); umwandla (Z) De St/sh; D(samara) EBENACEAE Euclea natalensis A.DC., SI 201 PRU (e) large-leaved guarri(e); hlangula(r); inkunzi-emnyama(z) Eg St/sh; F(berry) Used to treat general diseases and also as tooth sticks. ERYTHROXYLACEAE Erythroxylum emarginatum Thorn., SI 185 PRU common coca tree(e); uphaphane(z) Eg St/sh; F(berry) EUPHORBIACEAE Cleistanthus schlechteri (Pax.) Hutch. SI 194 PRU [21] bastard tamboti(e); umzithi(z) De Mt/st; D(capsule) Used to treat evil. Mixed with roots of Synaptolepis kirki, the decoction or infusion is taken three times daily. Croton gratissimus Burch., SI 240 PRU lavender-croton(e); ilabele(z) De St/sh; D(capsule) menyhartii Pax., SI 139 PRU rough-leaved croton(e) De St/sh; D( capsule) pseudopulchellus Pax., SI 130 PRU small lavender tree(e); dilhambo(r); uhubeshane(z) De St/sh; D(capsule) Used to treat malaria. Inhalations of infusions of the leaves are taken two times daily. 86

94 steenkampianus Gerstner, SI 179 PRU (e) marsh croton(e); uhubeshane omkhulu(z) De St/sh; D( capsule) Drypetes arguta (Mull Arg.) Hutch., SI 195 PRU water drypetes(e); iskolemfene(z) Eg St/sh; F (berry) natalensis (Harv.) Hutch., SI 200 PRU natal drypetes(e); umgunguluza(z) Eg Mt/st; F(berry) Hymenocardia ulmoides 0/iv., SI 191 PRU red heart-tree(e); tshatsalatani(r); umdlamahlathi(z) De M t/st; D ( samara) Sapium integerrimum (Hochst. ex.c.krauss) J. Leonard., SI 176 PRU duiker-berry(e); encheha(r); umhlepha(z) De Mt/st; D(capsule) Used to treat heart diseases and high blood pressure. Decoctions of the root are taken three times daily. Suregada zanzibariensis Bail!., SI 146 PRU woodland suregada(e); djlamunti(r); umdlankawu(z) St/sh; D(capsule) Used to treat fever. Decoctions of the leaves are taken. 87

95 FABACEAE Albizia adianthifolia (Schumach) W. Wright, SI 161 PRU rough-bark flat-crown(e);gouana(r); igowane(z) De Lt; D(pod) versicolor 0/iv., SI 212 PRU large-leaved false thom(e); umpiso(r); umphisu(z) De Lt/mt; D(pod) Afzelia quanzensis Welw., SI 209 PRU [25] pod mahogany(e); chanfuta(r); inkele(z) De Lt/mt; D(pod) Crotolaria monteiroi var. monteiroi Taub. ex. Baker f., SI 166 PRU (e) Small-leaved rattle-pod(e); lekalahumba(r); uhashazi(z) SDc Lt/mt; D(pod) Dalbergia nitidula Baker, SI 132 PRU glossy flat-bean (E) De Mt/sh; D(pod) Dialium schlechteri Harms, SI 129 PRU [18] zulu-podberry(e); insiba(r); umthiba(z) De Mt/st, D(pod) Used to treat chest pains. Roots, together with roots of Manilkara disco/or, are boiled and the decoction is taken three times daily. Erythrophleum lasianthum Corbishley, SI 147 PRU (e) swazi ordeal tree(e); umhlakazane(z) De Lt/mt; D (pod) Used as remedy for headaches, migraine and general pains. The powered bark is used as snuff, after being dried and pounded. Indigofera podophylla Benth ex.harv., SI 204 PRU (e) hlenyenyani(r) De St/sh; D(pod) 88

96 Mundulea sericea (Wild.) A. Chev., SI 163 PRU cork bush(e); citana(r); umamentabeni(z) SDc/Eg St/sh; D(pod) FLACOURTIACEAE Dovialis longispina (Harv.) Warb., SI 187 PRU (e) natal apricot(e); nhangatsune(r) Eg Mt/st; F(berry) Xylotheca kraussiana Hochst., SI 155 PR african dog rose(e); gutana(r); isishwashwa(z) Eg St/sh; D(capsule) Used as anthelmintic in childrens. Roots are boiled and the decoction is taken before meals, three times daily. LAMIACEAE Plectranthus sp., SI 230 PRU bozane (R) H/sh(P); D(nut) Used to treat fever. Leaves are boiled and the decoction is taken three times daily. LAURACEAE Cassytha filiformis L., SI 216 PRU false dodder(e); umkhunga(z) Eg H(P); F( drupe) Used to treat malaria. An infusion of the root, mixed with root of Croton pseudopulchellus, is taken twice daily. MALPIGHIACEAE Acridocarpus natalitius A. Juss var. linearifolius Launert, SI 162 PRU (e) moth-fruit(e); mabope(r); umabopha(z) S t! shl cl; D( samara) Used to prevent several illnesses. Roots are boiled and decoctions taken three times daily. 89

97 MELASTOMAT ACEAE Wameckea sousae (A. & R. Fernandes) A.E. van Wyk, SI 144 PRU (e) Tonga false rose apple (E) Eg St/sh; F(berry) MELIACEAE Pseudobersama mossambicensis (Sims) Verdc., SI 189 PRU false white ash(e); umopho(z) Eg Mt/st; D (capsule) MENISPERMACEAE Albertisia delagoensis (N.E.Br.) Forman., SI 228 PRU umgandaganda(z) Eg Sh; F(berry) Used to treat constipation, intestinal complications and worms. Cissampelos hirta (Lodd.) Engl., SI 196 PRU (e) djebehabo(r) Eg Cl; F(berry) Used as remedy for stomach problems, diarrhoea and cholera and also for easing childbirth. Roots are boiled and the decoction or infusion taken three times daily. MYRTACEAE Eugenia mossambicensis Engl., SI 134 PRU (e) nkelenkele (R); St/sh; F(berry) Used to prevent worms in children. Water of boiled roots is taken three times daily. OCHNACEAE Ochna barbosaen. Robson, SI 174 PRU (e) [24] sand plane(e); giwongwani(r) De St/sh; F( drupe) 90

98 natalitia (Meisn) Wa/p., SI 171 PRU (e) natal plane(e); driwongwani(r); isendengulube(z) De St/sh; F( drupe) Used to treat cough. Roots and leaves are boiled and the decoction is taken as a tea three times daily. OLACACEAE Olax dissitiflora Oliv., SI 199 PRU small sourplum(e); lijamuntani(r); umaphunzana(z) De St/sh; F( drupe) Used for fever conditions. Leaves are mixed with the bath water for children to decrease in high temperatures. OLEACEAE Olea capensis L., SI 234 PRU false ironwood(e); umsishane(z) Eg Lt/st/sh; F( drupe) europaea L. subsp. africana (Mill.) P.S.Green, SI 127 PRU wild olive(e); umnqumo(z) Eg Mt/st/sh; F( drupe) POLYGALACEAE Securidaca longepedunculata Fresen., SI 229 PRU violet tree(e); mulha-lhovo(r) Dc/Eg St/sh; D(samara) Used to treat cough, and chest complaints, toothache and headache. For chest complaints and rheumatism a decoction of parts of the plants is taken while the root is chewed to relieve toothache. The powered root or wood scrapings are rubbed into scarifications on the forehead to treat headaches. 91

99 PTAEROXYLACEAE Ptaeroxylon obliquum (Thunb.) Radlk, SI 178 PRU Sneezewood(E); dzarre (R); ubhaqa(z) De Lt/st; D( capsule) Used to treat headache, rheumatism and arthritis. Powered wood is used as a snuff to relieve headache. A decoction or infusion of the wood and bark is used for treatment of rheumatism and arthritis. RHIZOPHORACEAE Cassipourea mossambicensis (Brehmer) Alston, SI 150 PRU (e) sand onionwood(e); mnyamanzi(z) Eg St/sh; D( capsule) sp., SI 239 PRU urungwe onionwood(e) Eg St; D( capsule) RUBIACEAE Canthium setiflorum subsp. setiflorum Hiern., SI 172 PRU (e) rough-leaved rock(e); txandulankumba(r); umbhangwe(z) St/sh; F(drupe) Catunaregam spinosa (Thunb.) Tirvengadum, SI 211 PRU thorny bone-apple(e); balekane(r); umaqhathini (Z) De Mt/st; F(drupe) Used against worms in child. Roots decoction is taken three times daily. H yperacanthus amoenus (Sims) Bridson, SI 236 PRU spiny-gardenia; thorny gardenia(e) Eg St/sh;F( drupe) microphyllus (K. Schum.) Bridson, SI 145 PRU (e) [23] umtembezwana(z) Eg St/sh; F(drupe) Used to treat spirit problems. Roots are boiled with those of Tarenna supra-axilaris and taken three times daily. 92

100 Leptactina delagoensis K. Schum., SI 184 PRU sandnightstar, nightstarbush(e) Eg Sh; F( drupe) Oxyanthus latifolius Sond., SI 157 PRU (e) [22] Zulu loquat(e); umdlankawu(z) Eg St/sh; F(drupe) Used to treat abscesses. Fresh leaves are chewed and boiled and the infusion used externally. Pavetta gerstneri Bremek., SI 237 PRU (e) Zulu bride's bush(e); chibandichachoco(r) De St/sh; F(drupe) schumanniana F.Hoffm. ex. K. Schum., SI 141 PRU poison bride's bush(e); isimbuzana(z) De Sh; F(berry) Psydrax fragrantissima (K. Schum.) Bridson, SI 135 PRU (e) [17] Tonga quar(e) Eg Shlt; F( drupe) locuples (K. Schum.) Bridson, SI 138 PRU (e) [16] krantz quar(e); xixlovongo(r) Eg St/sh; F(drupe) Used to relieve bums. Dried chewed leaves are applied on wounds. obovata (Eck/.& Zeyh.) Bridson SI 214 PRU quar (E); terentele(r) Eg St/sh; F(drupe) 93

101 Rothmannia fischeri (K. Schum.) Bullock, SI 210 PRU woodland rothmania(e); umkotshi-wenyama(z) Eg St/sh; F(berry) Tarenna junodii (Schinz) Bremek., SI 156 PRU climbing dune tarenna(e); ligangani(r); amabamba(z) De St/sh; F(drupe) nigrescens (Hook f) Hiern., SI 238 PRU blackbutterspoon(e) De Cl/ sh/ st; F( drupe) supra-axilaris (Hems/.) Bremek., SI 183 PRU narrow-leaved false bride's bush(e); ukukulaketelo-lehlathi(z) De St/sh; F(drupe) Used against evil. Cold water is poured over a small quantity of powdered roots, and the patient takes the foam. Tricalysia delagoensis Schinz., SI 167 PRU (e) [20] tonga jackal-coffe(e); tsindi (R) De Shlst, F( drupe) junodii (Schinz.) Brenan var. junodii., SI 148 PRU (e) xisimelahlapfi(r) De Sh; F( drupe) lanceolata (Sond.) Burtt Davy SI 128 PRU common tricalysia (E); thisindi (R); umdleza (Z) Eg St/sh; F(drupe) maputensis Bridson & A.E. van. Wyk ined. SI 177 PRU (e) [19] Pe Sh; F( drupe) 94

102 RUTACEAE Toddalliopsis bremekampii 1 Verd., SI 151 PRU (e) wild mandarin(e); intane(z) Eg St/sh; F(berry) Used to treat rheumatism. Decoctions or infusions of the roots are taken three times daily. Teclea cf. myrei Exe/1 & Mendonfa, SI 235 PRU cherry orange(e); mutza (R) Eg St; F(berry) SAPOTACEAE Inhambanella henriquessii (Engl. & Warb.) Dubard, SI 188 PRU inhambanella(e); umbenkela(z) Eg T; F(berry) Manilkara discolor (Sond.) J.H.Hemsl., SI 160 PRU forest milkberry(e); nheve(r); umnqambo(z) Dc/SD c Lt/mt; F(berry) SCROPHULARIACEAE Buttonia natalensis Mckew ex. Benth., SI 232 PRU umhlanhalaze enkulu(z) Eg Shlcl(P); D( capsule) Used to treat general infections. Root decoction is taken three times daily. 95

103 STRYCHNACEAE Strychnos madagascariensis Poir., SI 153 PRU black monkey orange(e); macuacua(r); umnconjwa(z) De St/sh; F(berry) spinosa Lam., SI 131 PRU green monkey orange(e); massala(r); umhlalakolontshe(z) De St/sh; F(berry) THYMELAEACEAE Synaptolepis kirkii 0/iv., SI 140 PRU uvuma omhlophe(z) Eg Sh; F(berry) Used as an emetic. Also used to treat epilepsy and constipation. A root decoction or infusion is taken three times daily. TILIACEAE Corchorusjunodii (Schinz) N.E.Br., SI 180 PRU (e) chimbati (R); igusha(z) H/sh (P); F( capsule) VERBENACEAE Vitex amboniensis Guerke, SI 149 PRU large-fruited vitex(e); umphenduka(z) De St/sh; F(berry) Used to treat conjunctivitis and cataracts. Roots decoctions are used to wash the eyes, four days weekly. VITACEAE Rhoicissus revoillii Planch., SI 142 PRU bushveld grape(e); chilucarelo(r) De Cl; F(berry) The latex is used to treat cows cataracts and is applied to the eyes two times daily. Cissus rotundifolia (Forssk.) Vahl., SI 193 PRU bushveld grape(e); sungaunhe(r); umjovane(z) De Cl; F(berry) 96

104 ANNOTATED CHECKLIST LEGEND Scientific name, follows by Authority, sometimes by var. =variety or subsp. =Subspecies. Endemicity: (e)= Endemic/near-endemic to Maputaland Centre of Endemism. Common Name: E = English, R = Ronga, Z = Zulu. Life History: Eg =Evergreen, De= Deciduous, SDc = Semi-deciduous, Pe =Perennial. Growth form/habit: H =Herb, G =Grass, Cl= Climber, L = Liane, Sh =Shrub, ST = Smal Tree, MT =Medium Tree, LT= Large Tree, (P) =Parasite. Fruit Type: F = Fleshy, D = Dry; sometimes followed the by specific fruit type in brackets. Medicinal Uses 97

105 APPENDIX2 MISCELLANEOUS PHOTOGRAPHS 98

106 1. Mrs Izidine in Licuati Thicket during plant collecting field trip Mr Macitela (left), Mrs Izidine and Mr Manhique (right) investigating plants.

107 Chicken prepared for traditional ceremon) 4. Mr N guenha cutting chicken. 5. Mr Santaca pouring wine for the anceston

108 ~ 0 6. Group photo taken at Sacred Forest. ~ From left to right: Mr Macitela, Mr Vilanculos, Mrs Izidine, Mr Santaca, Mr Manhique and Mr Ngwenha. ersity of Pretoria

109 7. Aerial photograph oflicuati Forest patches in woodland, southern Mozambique. 8. Aerial photograph oflicuati Forest, southern Mozambique.. 102

110 9. Aerial photograph showing Licuati Thicket, Licuati Forest Reserve, southern Mozambique. 103

111 10. Road through Licuati Thicket, Licuati Forest Reserve. 11. Road through Licmiti Thicket showing undulating dunes.

112 12. Xylopia torrei (flowers). 13. Xylopia torrei (young fruit). 14.Xylopia torrei (dehisced fruit). 105

113 15. Combretum mkuzense (MC endemic). 16. Psydrax locuples (MC near-endemic). 17. Psydrax fragrantissima (MC endemic). 18. Dialium schlechteri (MC near-endemic).

114 19. Tricalysia maputensis (MC endemic). 20. Tricalysia delagoensis (MS near-endemic). 21. Cleistanthus schlechteri (fruit). 22. Oxyanthus latifolius (MC endemic). 107

115 23. Hyperacanthus microphyllus (MC endemic). 24. Ochna barbosae (MC near-endemic). I. 25. Aftelia quanzensis (Chanfuta) chopped down for timberwood. 108