Name, Location and Area

The Sango Bay Forest Reserve is located in southern Uganda near Lake Victoria and consists of Kaiso, Malabigambo, Namalala, Tero West, and Tero East forest blocks (Figure 1). The Kaiso and Malabigambo blocks are contiguous with the Minziro forest blocks in Bukoba District, Tanzania.

Figure 1

Location of the Sango Bay Forest Reserve (Data source: National Biodiversity Databank at the Makerere University Institute of Environmment and Natural Resources (MUIENR)).

There are three human settlements in the reserve: Minziro, Kanabulemu and Gwanda-Kasensero. The former two are enclaves in the forest reserve, while Gwanda-Kasensero is partly surrounded by the forest. The reserve is located between latitude 0°47 and 1°00′ S and longitude 31°28 and 31°43′ E. The total area of Sango Bay Forest Reserve is 576 km² of which approximately 180 km² is forested and almost 400 km² is grassland. The grassland areas now in the reserved forest lands were public lands until recently when they were gazetted.

The study was carried out in the villages of Kanabulemu and Minziro located in Kakuuto subcounty, Rakai district. There are over 150 households and a combined population of about 550 people. The area possesses unique physical, biological and ethnic features. The terrain is characterized by moderate to steep hill slopes. It experiences bimodal rainfall with peaks around April/May and October/November. The major ethnic groups include the Baganda, Banyankole, Barundi, Banyarwanda and the Baziba from northern Tanzania. These are mainly Christian peoples (about 80%), with few Muslims.

The inhabitants are mainly subsistence farmers living in small scattered settlements. Bananas, maize, cassava, sweet potatoes, and beans are the main food crops while coffee and, lately, beans are the main cash crops. There are migrant pastoralists who are mainly Banyarwanda and Banyankole. They graze their cattle in the vast grasslands that comprise part of the forest reserve.

The main diseases afflicting the local communities are gastro-intestinal diseases that primarily affect the children. However, malaria is the principal specific disease that causes the highest mortality rates, especially among children under the age of five years. In most cases, traditional medicine is used in the treatment of the diseases.

Vegetation Types

The Sango Bay Forest Reserve is a complex forest/wetland ecosystem. The vegetation of the five forest blocks was broadly classified as Baikiaea-Podocarpus seasonal swamp by Langdale–Brown et al. (1964). However, significant changes have taken place in terms of mature ecosystem species composition, as a result of several years of commercial logging. Recent surveys by the National Forest Authority indicate that the current canopy species belong predominantly to the genera Baikiea and Cleistanthus. It is the only site in East Africa for this specialized community where Afrocarpus dawei (Stapf) C.N. Page and Podocarpus milanjianus Rendle are known to grow at such a low altitude. The grasslands, which form some two-thirds of the reserve, are predominantly tussocks of Miscanthus violaceus (K. Schum) Pilg., Loudetia kagerensis (K. Schum.) Hutch., and Themeda triandra Forssk. The tussock-like nature of the grasslands is a result of seasonal flooding accompanied by erosion. Syzygium cordatum Hochst. is the major tree species along wetter drainage areas.

Ethnobotanical Survey

Field work for this study was carried out between January and October 2003. Semi-structured interviews, questionnaires, and direct observation were the main forms of data collection. Two local assistants helped in establishing rapport with the local communities and with data gathering. One of the assistants was a forest ranger and had considerable knowledge of the forest.

A pilot study over one week was conducted at the beginning of the study to gather a general understanding and test the questionnaire. Following the pilot study, a detailed survey using a mixture of open-ended and close-ended questionnaires during face-to-face interviews was conducted. Interviews were conducted in the local language, the Luganda (Buddu dialect). Household respondents were selected randomly, although we never interviewed anyone under 18 years old because they were presumed to have limited knowledge of traditional medicine, and did not generally partake in the preparation and administration of medicines. We interviewed either the head of the household, wife or older children. In some cases we interviewed both the head of the household and his wife, if during the course of the interview we discovered that the wife had good knowledge of medicinal plants (possibly served as a traditional herbalist). Effort was made to get key informants, either in the form of traditional herbalists (n  =  2 male) and the elderly respected people (n  =  3 male/ 1 female) to verify the information gathered from the households. Informal discussions were also held with the local people to get additional information. This would help verify and corroborate information gathered (Briggs 1986; Crane and Angrosino 1992).

During the interviews, respondents were asked to rank the species in order of importance in the treatment of malaria, regardless of their abundance or availability. This would help identify the three key forest tree species considered most important by the people in the treatment of malaria. These were studied further to assess their ecological viability (see below). Altogether, 102 respondents (63 males and 39 females) from 69 households in the villages of Kanabulemu and Minziro were interviewed in the survey. Eighty respondents (78.4%), 54 male and 26 females, acknowledged that they use traditional medicine to treat malaria. The questions in the questionnaire included what plants they used against malaria and the parts they used, frequency of use, and local/traditional plant management measures. The interviews were supplemented by direct observations and walks. During the walks, plants used by the people were identified using local languages and scientific names ascertained from voucher specimens.1

Plant Population Structure: Layout of Transects and Sample Plots

Plots of 20 × 50 m (0.1 ha.) were used to census trees for diameter at breast height (DBH), saplings, poles, and mature trees of the three key species in 45 plots (Tchouto 2004). Recording of seedlings was also done in the plots, and presence of mother trees called for a more rigorous search for the seedlings. It was relatively easy to census the seedlings because we targeted the key species only. Stems in the plots were categorized in different classes up to over 60 cm DBH. Plots were located along transects that varied from 1400 m to 2000 m at intervals of 250 m. Diameter of each tree was measured at breast height (1.3 m above the ground) using a diameter tape. Diameters of trees with buttresses were measured at a point just above the buttresses. For each species, seedlings (diameter < 5 cm), and diameter classes of (5 – 19.9 cm), (20 – 39.9 cm), (40 – 59.9 cm) and (DBH > 60 cm) were identified and counted, and diameter was determined. There were no stumps of the target species that could be identified.

To provide an indication of the degree of exploitation, and therefore intensity of harvesting of bark or mutilation, the debarked section on the stem of each mutilated tree was categorized. The extent of debarking was determined in relation to diameter and size of debarked section of the tree (1 =  small section of stem debarked; 2 =  big section of stem debarked; 3 =  very big section of stem debarked) (see Dhillion and Gustad (2004) for this approach). To minimize errors in identification, voucher specimens for the lesser known species were collected and each was given a specific number.

Data Analysis

Ethnobotanical data was analyzed both qualitatively and quantitatively. Responses from the open-ended questions were grouped into classes that expressed similar ideas while percentages, based on valid responses only, were calculated from close-ended questions. Selection of the most important tree species used to treat malaria was based on the frequency of reporting and preference ranking, as used by Noweg et al. (2003). The Friedman test (Ludwig and Reynolds 1988; Zar 1984) was used to determine whether there were significant differences in the use of plants by respondents. We evaluated for significant differences in use of medicinal plants to treat malaria with education level and marital status through the use of the non-parametric Kruskal-Wallis one-way analysis of variance test.

Cluster Analysis: Similarities in the Use of Species

We used cluster analysis to sort objects (such as sampling units) into groups or clusters based upon their overall resemblance to one another (Ludwig and Reynolds 1988). To place species into meaningful groups, the Simple Matching Coefficient (SM) for all combinations of site pairs were summarized into a 16 × 80 similarity matrix. An agglomerative clustering technique (weighted centroid) provided in the Multivariate Statistical Package of Kovach (1999) was used to produce a dendrogram containing all 16 species.

A minimum SM of 0.2 was used for defining clusters. Other measures of similarity and also measures of distance between sites were attempted along with several different methods of clustering (single-linkage and complete-linkage techniques). All techniques provided similar results, with the SM and weighted centroid clustering being most meaningful from the ethnoecological point of view.

Significance of Clusters

To determine whether there were significant differences between the use of the different species and also whether there was significant differences between the clusters of species generated using cluster analysis, the Friedman test was used.

Socio-demographic Attributes of the Respondents

The majority of the respondents (67.5% of respondents) were males and had basic primary level education (83.8% of the respondents) as shown in Table 1. Seventy-five percent of respondents were under 38 years of age and the majority were married (88.8% of the respondents). Their income was so low that the majority were characterized as poor (66.3% of the respondents). Being poor meant that you could earn about US$ 1 per day, living in temporary houses constructed from mud and thatched with grass (26.3% of the respondents).

Table 1

Socio-demographic attributes and traditional medicinal plant management practices of the respondents.

Since the majority of them were married, this can imply that they may have children who may fall sick of malaria and/or related ailments. This necessitates the parents, in particular the head of the household, to look for the medicinal plants in the forests and adjacent grasslands as the wives do the domestic chores. Neither the level of education (K  =  0.782, df  =  2, P > 0.05 Kruskal-Wallis test) nor marital status (K  =  1.091, df  =  2, P > 0.05 Kruskal-Wallis test) seemed to influence the choice of use of medicinal plants against malaria, but rather the need to treat malaria.

Medical Practices

Most of the respondents used malaria medicinal plants two to four times per month (67.5% of the respondents), traveling over 1 km (57.5% of the respondents) in search of the medicinal plants in grasslands, abandoned fields or forest (Table 1). However, the health centers were reportedly further from residential homes, with a majority reporting that they are over 1 km away (70% of the respondents).

Many of the respondents argued that they would only go to health centers if the traditional medicine failed (87.8% of the respondents), while others would only go there with acute cases of malaria because traditional medicine does not act as quickly as modern medicine (Table 1).

Traditional Medicinal Plant Management

Very few respondents cultivate medicinal plants used to treat malaria (Table 1, 6.6% of the respondents). Cultivation was mainly practiced by the few traditional herbalists. However, many preserved those medicinal plants that grow in home gardens and abandoned fields (47.1% of the respondents). These were primarily herbaceous medicinal plants, as shown in Table 2. There was a tendency to nurture valued but rare medicinal plant species, especially in the wooded grasslands adjacent to the forest, such as Warburgia ugandensis (25.1% of the respondents).

Table 2

Species used to treat malaria by the local communities showing their respective importance through ranking. Gr  =  Grassland, SF  =  Secondary fallow, Fo  =  Forest, Ga  =  Garden.

Plants Used to Treat Malaria

During the ethnobotanical survey, 16 plant species were reported by the respondents to treat malaria (Tables 2 and 3). These represented 11 families and 14 genera. Four species belonged to the family Fabaceae. Asteraceae and Myrtaceae were represented by two species each, whereas the other families were represented by a single species each. Eight species (50%) were trees, two species (12.5%) were shrubs and six species (37.5%) were herbaceous plants. Eight (50%) of the species recorded were typical forest species, whereas the others also occur in grassland and other vegetation types. The most important species used was Hallea rubrostipulata (tree) ranked by 51 (63.8%) of the respondents followed by Vernonia amygdalina (shrub), Warburgia ugandensis (tree) and Syzygium guineense (tree) being ranked by 40%, 28.8% and 16.3% of the respondents, respectively. Only tree species were chosen for ecological viability studies, so Vernonia amygdalina, a shrub, was excluded.

Table 3

Methods of preparation and application of medicinal plants used to treat malaria

Local Use and Importance of Medicinal Plants

Hallea rubrostipulata, Warburgia ugandensis and Syzygium guineense were ranked as the most important forest tree species used in the treatment of malaria (Table 2). The majority of the plants have been reported in earlier studies to treat malaria and other ailments (Tables 2 and 3). Vernonia amygdalina (a shrub) has been reported to be widely used in the country to treat malaria and other ailments (Adriaens 2005; Bukenya-Ziraba et al. 1997; Kakudidi et al. 2000; Katende et al. 1995, 1999; Ogwal-Okeng 1998; Oryem–Origa et al. 2003; Tabuti et al. 2003b). This study reports the use of four species for the first time: Hyptis pectinata Poir., Indigofera congesta Baker, Manilkara obovata (Sabine & G.Don), and Sopubia ramosa (Hochst.) Hochst.

The majority of respondents used more than one plant for malaria treatment (60%). Some respondents used up to four plant species as a combination (6.3%), although the majority used two species (35%) and others three (18.8%). The most frequently used combinations of plants included at least two of the following plants: Pseudarthria hookeri Wight & Arn., Indigofera congesta Baker, and Senna didymobotrya (Fresen) Irwin & Barneby. The other combinations involved Pseudarthria hookeri Wight & Arn., Hyptis pectinata Poir., Entada abyssinica Steud. ex A.Rich. including Erythrina abyssinica Lam. ex DC.

Plant Parts Used and Methods of Application

The bark was the most widely used plant part, accounting for 37.5% of the plant uses (Table 3). A majority of the remedies were prepared in the form of decoctions from the bark; the bark was collected and boiled in water (Table 3). All the remedies were taken orally. To improve the acceptability of certain remedies, they were mixed with an additive: sugar could be added to Vernonia amygdalina infusion to reduce the bitterness; lemon (Citrus limon) was added to Senna didymobotrya to improve the taste; or the powder of the bark of Warburgia ugandensis was placed in a piece of sweet banana and swallowed. Sometimes the contents of orthodox medicine drug capsules bought from drug shops were removed and replaced with the powder of the bark of Warburgia ugandensis and swallowed. This was done as a measure of ‘dosage’, and also because W. ugandensis has a hot peppery taste and, if taken in an ‘overdose’, can cause stomach upsets, vomiting and diarrhea.

In some parts of the country the same plant parts of a species, as recorded in this study, were also used to treat malaria. For example, according to Adriaens (2005), the bark powder of Warburgia ugandensis was prepared and administered in the same way as the people of the Rwenzori region in western Uganda (Table 3). In some instances, the same or different plant parts of a particular species, as recorded in this study, were used to treat a different ailment. This is evident, for example, with the use of the leaves and bark of Senna didymobotrya to treat stomach ache, asthma, measles and as a purgative in western Uganda (Table 3), as reported by Kakudidi et al. (2000) and Adriaens (2005). The root infusion was also used to treat diarrhea in Eastern Uganda (Tabuti et al. 2003b).

Cluster analysis: Similarities in the Use of Species

Using a minimum simple matching coefficient of 0.2 as an arbitrary cut-off point for defining clusters (Figure 2), three species assemblages were produced by the cluster analysis. Assemblage A included mainly those species that predominantly grow in seasonally flooded wooded grasslands, farmland, and disturbed areas, and are fairly common. The plant parts used were generally bark or the entire plant. Assemblage B had those species that are predominantly herbaceous and grow in disturbed areas, gardens and secondary regrowth. The main plant part used were whole plant or leaves. Assemblage C included those species that are typically forest tree species and main plant material used was the bark.

Figure 2

Dendrogram based on the similarity matrix and showing clustering of 16 species used to treat malaria according to the respondents. The species are Senna didymobotrya (Sendid); Hyptis pectinata (Hyppec); Syzygium cordatum (Syzcor); Entanda abyssinica (Entaby); Zanthoxylum chalybeum (Zancha); Pseudarthria hookeri (Psehoo); Sopubia ramosa (Sopram); Shirakiopsis elliptica (Shiell); Manilkara obovata (Manobo); Indigofera congesta (Indcon); Justicia anselliana (Jusans); Vernonia amygdalina (Veramy); Vernonia lasiopus (Verlas); Syzygium guineense (Syzgui); Warburgia ugandensis (Waruga); Hallea rubrostipulata (Halrub). Species assemblages (A – C) were defined using a simple matching coefficient of 0.2 (dashed line).

The Friedman test (Zar 1984) showed that there was a highly significant difference between the use of species by respondents and also between clusters χ² _{.001 [2]}  =  21.393, P < 0.001 (Figure 2). In terms of the species present within each cluster, these were considered unique from all others for analyses of the respondents associations. Cluster C constituted the most important forest tree species used to treat malaria (Table 2). They also had a similar plant part harvested (bark) and are typical forest species, hence a subject of further investigation in this study.

Tree Density of the Key Species used to Treat Malaria

Hallea rubrostipulata was the most abundant species and Warburgia ugandensis was the least abundant. Hallea rubrostipulata had the highest mean density of 14.8 stems ha⁻¹ (SD  =  6.534) while Syzygium guineense had the least mean stem density of 8.4 ha⁻¹ (SD  =  4.277). The number of mature individuals (≥ 40 cm DBH) per sampled plot ranged from zero in Warburgia ugandensis to two in Hallea rubrostipulata. Very few (4.1%) sampled plots had more than one stem of mature target tree species.

Diameter Size Classes of the Tree Species

The frequencies of tree species in each DBH class from the forest are presented in Figure 3. The diameter size classes of Hallea rubrostipulata, Syzygium guineense and Warburgia ugandensis generally show a similar pattern. They represent a ‘reverse J-shaped’, suggesting a continuous recruitment. Three size classes defined were: (DBH < 10 cm), medium (DBH 10 – 40 cm) and large (DBH > 40 cm).

Figure 3

Number of individuals per hectare (ha) in a range of diameter size classes measured at diameter at breast height (dbh) for Hallea rubrostipulata, Syzygium guineense and Warburgia ugandensis.

Degree of Mutilation through Exploitation by Debarking

It can be seen that there was a higher degree of mutilation of Hallea rubrostipulata (63.8% of the respondents) for malaria treatment as compared to the other species as shown in Table 2 and Figure 4. Hallea rubrostipulata also had the highest stem density of 14.8 stems ha⁻¹. Warburgia ugandensis was the least mutilated, and therefore, least utilized as a medicinal tree species for the treatment of malaria and also had the lowest stem density. According to the local communities, these species were also increasingly being harvested for commercial purposes, particularly Hallea rubrostipulata and Warburgia ugandensis.

Figure 4

The degree of mutilation of trees expressed by percentage (%) of stem debarked of the most important forest tree species used to treat malaria.

Habitat, Current Status and Trend of Medicinal Plants

This study shows that there is little practice of cultivating medicinal plants in the area (6.5% of respondents), as shown in Table 1. Most of the medicinal plants were, therefore, harvested from the wild. A few species, for instance, Sopubia ramosa, Indigofera congesta, Hyptis pectinata and Vernonia amygdalina, were preserved in home gardens or farm plots for medicinal purposes (47.1% of the respondents). Most of the species were collected from the forest or adjacent grasslands at varying distances from residential homes (Table 1). Warburgia ugandensis is so rare in the forest that only those people who know the exact locations of these few individuals are normally sent to collect the bark material. Most of the herbaceous medicinal plants are commonly or occasionally found in the area. Most of them are wild herbs that grow in disturbed areas such as secondary vegetation, abandoned farmland, roadsides and around home gardens (Table 2). Fuelled by the continually increasing habitat disturbance, the abundance of some of the plants is reportedly increasing over years.

People, Their Dependence on and use of Traditional Medicine to Treat Malaria

This study clearly shows that the reliance on medicinal plants to treat malaria by the local communities is high, and that a number of species are used. Reliance on traditional medicine has been demonstrated in earlier studies, but this study further highlights the importance of traditional medicine in the daily lives of the people of Sango Bay, and the preferences for particular species over others in the treatment of malaria in southern Uganda. Some of these species are reported for the first time. Shrestha and Dhillion (2003) report that such significant attachment to the use of traditional medicine can be attributed to, among other reasons, the traditional belief in its effectiveness.

Relevance of Traditional Medicine in the Treatment of Malaria

Sixteen species of plants were recorded to be used to treat malaria. These occur in different habitat conditions and have varying life forms. The diversity of species used for the treatment of a single disease, malaria, can be an indication of the importance of these medicinal plants in primary health care. It also highlights the significance of malaria as a disease that afflicts both young and old, just as HIV/AIDS. Kakudidi et al. (2000), Tabuti et al. (2003b) and Oryem-Origa et al. (2003) reported over fifteen species used in the treatment of malaria in different parts of the country.

The reliance on the use of traditional medicine can be attributed to the perceived potency of these plants in the treatment of malaria, particularly for Hallea rubrostipulata. The absence of multiple health centers means that people must travel longer distances to health centers than to collect traditional medicine. This probably influences their choice of preference for traditional medicine to Western/orthodox medicine. In many cases, the health centers lack drugs that could be used. Therefore, people resort to the use of traditional medicine and only go to health centers when traditional medicine fails (Table 1).

Some species were reportedly used in combination with other species. Combining different species is believed to enhance their potency, especially among those known locally to be less potent. However, species including Hallea rubrostipulata and Vernonia amygdalina were never used in combination because they are known to be effective even when used individually.

Medicinal Plants and Associated Knowledge

Hallea rubrostipulata, Warburgia ugandensis and Syzgium guineense are the most important tree species used to treat malaria in the area. The high consensus on the use of traditional medicine against malaria shows the importance of these plants to the people. The high salience or familiarity of Hallea rubrostipulata, Warburgia ugandensis and Syzygium guineense (cited by 63.8%, 28.8% and 16.3% of the respondents, respectively) for use as remedies against malaria might indicate their efficacy. According to Trotter and Logan (1986), pharmacologically effective remedies are expected to have greater respondents' consensus. According to Johns et al. (1990), however, confirmation is not a single true measure of the potential efficacy of any remedy. There is great probability of a common plant, reported to treat a common disease, to be cited more frequently that a rare plant that is used to treat a disease of limited occurrence.

Most of the reported plants are also used elsewhere in Uganda for their medicinal value (Table 2). Eight species (Hallea rubrostipulata, Vernonia amygdalina, Warburgia ugandensis, Syzygium guineense, Zanthoxylum chalybeum, Justicia anselliana, Shirakiopsis elliptica, and Senna didymobotrya) are in many instances used in the same way as they are used by the local communities in Sango bay (Kakudidi et al. 2000; Tabuti et al. 2003b). The fact that some traditional medicine is used for the same purpose by more than one community might indicate the pharmacological effectiveness of these remedies.

The fact that four additional species have been reported for the first time as anti-malarial medicinal plants highlights the significance of malaria as a disease, among the people and the role of traditional medicine in primary health care of the local people of Sango Bay area.

Densities of the Most Important Tree Species Used to Treat Malaria

The results of the botanical survey show that the target species were among those species categorized as rare, with the exception of Syzygium guineense. Hallea rubrostipulata and Warburgia ugandensis occurred in 6.3% of the plots, whereas Syzygium guineense occurred in 52% of the plots. Rarity in forest ecosystems is a natural phenomenon, as most species are categorized as rare and very few are abundant or dominant (Magurran 1988). This is likely to influence the access and use of these species to treat malaria. The destruction of plants and extreme opening of canopy gaps often stimulate growth of dense, herbaceous and semi-woody intertwined vegetation that restrains tree regeneration (Omeja et al. 2004). This is true in this forest where the harvesting of Afrocarpus dawei has been going on until recently. This can probably explain the general poor regeneration of the species.

Although Syzgium guineense is relatively more abundant than Hallea rubrostipulata and Warburgia ugandensis, this study shows that Hallea rubrostipulata is more often used than Warburgia ugandensis. This could be related to the perceived potency of the different species in which Hallea rubrostipulata is more potent than the others. Potency of a traditional medicine can be a crucial factor in determining the relative use and importance attached to a species (Bukenya-Ziraba et al. 1997). Although there was evidence to suggest that these plants are being preserved in home gardens (Table 1), this is done at a very minimal scale. There is need to have them propagated in nurseries and encourage people to plant them. This is due to methods and intensity of exploitation threatening the populations of these species in the forests. In all the three species, the bark is harvested and several individuals had been excessively debarked. The use of alternative species, especially from the adjacent grassland, fallow land and abandoned fields, to treat malaria also eases the pressure on these species as observed during this study (Table 2).

Regeneration Characteristics

Unlike Hallea rubrostipulata and Syzygium guineense, the regeneration characteristics of Warburgia ugandensis are relatively poor compared to the other target species because of relatively fewer individuals in the seedlings stage. Although plenty of fruit was observed below some individual trees, there were very few seedlings in the vicinity. Studies have shown that forest disturbances, such as tree mortality, logging, alterations in tree phenology, and modifications in plant-animal interactions such as seed dispersal, can have a profound effect on the regeneration characteristics of a species (Bierregaard et al. 1992; Chen et al. 1992; Didham et al. 1996; Esseen 1994). When disturbances reduce animal dispersal, plants that depend on animals to carry their seeds will be affected. This results in the failure to ensure the regeneration of plants on other potential sites (Primack 1992). Whitmore (1991) also argues that many tropical tree species have poor dispersal capacities and hence are unable to ensure good regeneration.

Although Warburgia ugandensis was not represented by any individuals with over 40 cm DBH, there is good recruitment into the sapling and pole stages. It is probable that it rarely grows to attain a diameter over 50 cm naturally, although it grows up to a height of 25 m (Eggeling and Dale 1952; Katende et al. 1995). Hallea rubrostipulata and Syzygium guineense are represented in diameter classes over 40 cm DBH. This implies that there are enough mature reproductive individuals providing seeds for regeneration.

Human Exploitation of Key Plant Species through Debarking

Hallea rubrostipulata was the most utilized among the three species, probably due to its relatively high abundance. Harmful harvesting may affect the physiology and vital rates of individuals, change demographic and genetic patterns of populations, and alter community- and ecosystem-level processes (e.g., Dhillion and Amundsen 2000; Dhillion and Gustad 2004; Peters et al. 1989). More of the bark of Hallea rubostipulata is harvested compared to that of the other species, and therefore experience higher harvesting intensity. Another probable reason is that it is locally perceived to have higher potency compared to either Warburgia ugandensis or Syzygium guineense, as argued by Bukenya-Ziraba et al. (1997), for important medicinal plants recognized in a community.

Local Conservation Initiatives: Habitat and Status of Traditional Medicine

Most of the plants utilized by the local people are obtained from the wild. This is also true in many other parts of the country (Bukenya-Ziraba et al. 1997; Kamatenesi-Mugisha and Oryem-Origa 2005; Tabuti et al. 2003b). Some of the plants are preserved in the home gardens, such as Hyptis pectinata, for their medicinal value. The fact that most of the remedies are only found in the wild poses a significant threat to their existence if the destruction of their habitats continues. Continued growth of farmland, without leaving land to fallow because of shortage of land, can also be a threat to the continued use of herbal medicinal plants growing in agricultural fields.

Analysis of data collected from the forest shows that most of the forest medicinal tree species are rarely encountered. All the three species are typical forest species. They mainly occur in the forest and are rare even within the forest. However, we observed one individual of Warburgia ugandensis in farmland near the forests. It had been overly debarked but was still standing alive. However, no tree species were cultivated in the gardens except for the herbaceous species. This probably implies that the supply from the forest is sufficient to meet current needs, or the ecological conditions in farming systems do not allow the natural regenerations of the species. However, the continued degradation of the forest cannot guarantee the survival of these species. Therefore, this calls for the management regimes geared towards the protection of specific species such that they ensure a continuous supply of materials to use to treat malaria. This is being encouraged in other forests (e.g., in the Mpanga forest, Omeja et al. 2004) through a Collaborative Forest Management program.

Suggestions and Perspectives for Further Research

However, it should be noted that in many situations, the preparation methods of herbal medicine are not hygienic and need to be controlled to ensure that the medicines are safe to use. The therapeutic claims of these herbal medicines have not yet been evaluated for efficacy. These claims must be validated through further studies to validate effectiveness and safety of use of these plants as antimalarials by analyzing extracts for active compounds alongside toxicity determination. A good understanding of the active components would allow the development of extraction procedures and standardization of doses for treatment. This would raise self-assurance among the people who use the traditional medicine, as suggested by Tabuti et al. (2003b). This would also help create markets for some of the malaria herbal medicines. The marketing of herbal medicines would require detailed assessment of resource quantities, domestication possibilities, evaluation of market potential for promising species and setting up of equitable benefit sharing regimes (Dhillion and Amundsen 2000; Shrestha and Dhillion 2003; Tabuti et al. 2003b). There is need for continued monitoring of the harvesting regimes, as well as assessing the abundance and distribution of the key malaria medicinal plants so as to ensure sustainable exploitation of the plants.

The government of Uganda also recognizes that malaria continues to be the major cause of ill health and killer of Ugandans, followed by HIV/AIDS. Government, through the Ministry of Health, has put in place measures to combat the disease. These include formulation of a malaria treatment policy to enable the population at risk of malaria access to safe, good quality, effective and acceptable antimalaria drugs in order to: (1) ensure a rapid and long lasting clinical cure for individual malaria patients; (2) prevent progression of uncomplicated malaria to severe disease and death; (3) shorten clinical episodes of malaria and reduce occurrence of malaria-associated anemia in high malaria transmission areas; (4) reduce consequences of placental malaria infection and maternal malaria-associated anemia through chemoprophylaxis or intermittent preventive treatment during pregnancy; and (5) delay development and spread of resistance to antimalaria drugs.

The government has also developed various strategies intended to counter the malaria disease. These include the Home-Based Management of Fevers (HBMF) strategy aimed at increasing awareness about malaria symptoms and danger signs. It has helped caretakers and volunteers in taking appropriate action to treat malaria and increased access to treatment within 24 hours. The HBMF primarily takes place at community level and involves several key components: sensitization of members of the community and leaders; selection of community volunteers and mobilizers; and capacity building through training of volunteers and provision of anti-malarial drugs to volunteers and supervisors. The strategy aims at faster delivery of treatment and reducing the economic and human cost of malaria.

Mosquito-proofing of houses was one of the key methods for vector control in the 20^th century (Fullerton and Bishop 1933; Hackett and Missirolli 1932) and it should be effective against malaria in Africa since the majority of people receive infective bites indoors at night. Recent studies provide credible evidence that house screening is associated with protection against malaria transmission, infection and morbidity (Lindsay et al. 2002). Screening has the added advantage of protecting everyone in the room, and avoiding issues of discrimination and bias within the household. A pilot study conducted in the Gambia using experimental huts established that netting ceilings reduced exposure to malaria vectors by 80% (Lindsay et al. 2003), but more studies in actual houses are necessary to authenticate these findings and to assess the acceptability of house screening to communities.

Due to increased resistance of malaria parasites to chloroquine and sulfadoxine/pyrimethamine, a new and highly effective drug called artmether/lumefantrine (COARTEM) has been introduced as the first-line treatment for uncomplicated malaria, and artesunate/amodiaquine combination (ACT) as the alternative. In addition measures such as early treatment, using insecticide-treated nets and improved environmental management, the government intends to use indoor residual spraying with approved insecticides. The strategies outlined above, coupled with validated traditional medicine used to treat malaria, will contribute to increased control, and perhaps the reduction, of the malaria problem in Uganda.