Chapter 1. Introduction
Bats belong to the Order Chiroptera. At present there are about 950 recognized species worldwide, making up almost a quarter of all known mammal species. Bats are divided into two suborders, the Megachiroptera and the Microchiroptera. The Megachiroptera (consisting of a single family, the Pteropodidae or Old World fruit bats) are found throughout the Old World tropics and sub-tropics from Africa through southern Asia to Australia and on islands in the Indian and western Pacific Oceans. Microchiropteran bats are found in all areas of the world apart from the Arctic and Antarctica and some isolated oceanic islands.
The Megachiroptera are distinguished from the Microchiroptera by having a simple external ear with its edge forming an unbroken ring and by having a second finger that is relatively independent of the third finger and which usually bears a small claw. They do not possess any noseleaf (often well-developed in the Microchiroptera) or tragus (a small structure inside the ear). The tail membrane is usually narrow and the tail generally short or absent, although it is moderately long in Notopteris. Where a tail is present, it is not integral with the tail membrane.
The eyes are generally large, and sight and smell appear to be the major locational senses, in contrast to the Microchiroptera, which have small eyes. Echolocation, a method of orientation using ultrasonic sounds emitted through the nose or mouth, is universal among the Microchiroptera but is, with a few exceptions, unknown in the Megachiroptera. Where it is present (in some Rousettus and perhaps Epomophorus species), the acoustic orientation signals produced are rather crude and are made by a different mechanism from those in the Microchiroptera.
All of the Megachiroptera consume fruits, flowers and/or flower products. The grinding teeth of most species are large and flat to allow them to chew fruit. Nectar and flower feeders have relatively lighter jaws and smaller teeth, and usually have narrow, elongated muzzles and long tongues to allow them to probe deep into flowers. The majority of the Microchiroptera are insect feeders, although other food sources include fish, amphibians, small mammals (including bats), blood, fruit and flowers. In the Family Phyllostomidae the majority of species feed on fruit, nectar and pollen. It is suggested that this behaviour was derived from an insectivorous ancestor; there is no suggestion of such a derivation among the Megachiroptera (Hill and Smith, 1984).
The origin and evolution of bats is poorly understood. The earliest fossil bats come from the Eocene (approximately 60 million years ago), and were fully developed fliers. Thus, there is little information on the transition from their terrestrial ancestors. The Megachiroptera are first represented in the fossil record in the Oligocene (35 million years ago). All of the Eocene fossils are well-developed microchiropterans and undoubtedly do not include the ancestors of the Megachiroptera. It has been suggested that the Megachiroptera and Microchiroptera do not have a common ancestor and that wings and flight have developed twice, independently, in these two groups (Smith and Madkour, 1980; Pettigrew, 1986). Pettigrew et al. (1989) and Pettigrew (1991) suggested that the Megachiroptera evolved from an early branch of the primate lineage and that the Microchiroptera probably evolved much earlier from small, agile insectivores whose forelimbs had long metacarpals in relation to the phalanges. Megachiroptera share with primates a variety of complex details in the organization of neural pathways, which have not been found in any other mammalian group, particularly not in Microchiroptera (Pettigrew et al., 1989). This interpretation is challenged by Baker, Novicek and Simmons (1991).
The single megachiropteran family, the Pteropodidae, ranges from Africa, the eastern Mediterranean, Madagascar and the Indian Ocean islands in the west, across mainland southern Asia, throughout the islands of the western Pacific from the Ryukyu Archipelago and Ogasawara-shato in the north, to coastal eastern Australia, New Caledonia and the Loyalty Islands in the south, and east to Fiji, Tonga, Samoa and the Cook Islands (Figure 1). There are 41 genera containing a total of 161 species (K. Koopman, pers. comm.). The largest and best known genus, Pteropus, with 57 species, is primarily an island taxon, with 55 species (96.5%) having some or all of their distribution on islands. In this genus levels of endemism are extremely high, with 35 species (61.4%) confined to single islands or small island groups. Only nine species are found in continental areas (five in Asia and four in Australia), and only two (P. lylei and P. poliocephalus) are restricted to continents.
Figure 1. World distribution of the Old World fruit bats. Fruit bats have been recorded from within the hatched area.
The habitats used by fruit bats vary. Many taxa are dependent to a greater or lesser extent on primary or well-regenerated secondary forest. A few utilize areas of savannah. While forest destruction or degradation threaten many taxa, a few appear to accommodate such activities. For example, in the Philippines, Pteropus hypomelanus cagayanus and Rousettus (Rousettus) amplexicaudatus are most commonly found in disturbed habitats. Bats sometimes use habitats in large conurbations, as shown by the colony of Eidolon helvum in the campus of the University of Ife in Nigeria (Halstead, 1977). Some taxa are found in a great variety of habitats. For example, Cynopterus brachyotis has been recorded from montane forest, gardens, mangroves and strand vegetation in Borneo.
There have been few attempts to estimate population densities. Heideman and Heaney (1989) estimated densities of between 0.2 and 3.7 individuals per hectare for six small-bodied species on Negros.
Knowledge of roosting behaviour is fragmentary (Pierson and Rainey, 1992). A. G. Marshall (pers. comm.) has examined the published information on roosting sites of Megachiroptera. Twenty nine out of 41 genera roost in trees, 11 roost in caves, and six in various other sites (under eaves, in mines, rock shelters, crevices, buildings and amongst boulders). There is no information available for 10 genera. In trees, roost size varies from one to greater than 1 million, while in caves groups of between ten and several thousand have been found. Members of the genus Pteropus often form large aggregations on exposed tree branches. Large emergent trees such as banyan (Moraceae: Ficus benghalensis) are frequently used. Bats that roost singly or in small groups may use a variety of sites, such as crowns of epiphytic ferns or old termite nests in trees (Balionycteris maculata seimundi in Peninsular Malaysia), under dead palm leaves (Cynopterus brachyotis brachyotis in the Philippines), rock shelters (Cynopterus horsfieldii persimilis in Borneo), and in tree hollows or aerial roots of banyan (Cynopterus sphinx gangeticus in India). Cave-roosting bats may be found in light areas close to the entrance (Penthetor lucasi in Peninsular Malaysia, Eidolon helvum in Madagascar (Wilson, 1987)) or in the darker areas (Rousettus (Rousettus) lanosus kempi in East Africa). In Africa, Rousettus (Lissonycteris) angolensis roosts in small loose groups in cave entrances and cave-like habitats, and under dead palm fronds hanging down the sides of palm stems (Bergmans, 1979).
Agroforest habitat utilized by Pteropus insularis on Chuuk, Federated States of Micronesia. (Photo by W. E. Rainey)
Roost site fidelity is generally high in those genera that roost communally. Thus, cave roosts of Eonycteris, Notopteris and Rousettus may be occupied for many years (Marshall, 1983) as may tree roosts of Eidolon, Epomophorus and Pteropus (Rosevear, 1965; Lim, 1966; Funmilayo, 1976; Wickler and Seibt, 1976; Marshall, 1983). Those genera roosting singly or in small groups show less site fidelity but may use the same perch for considerable periods (Start, 1974; Marshall, 1983). Tree-roosting bats obtain protection from inclement weather and from predation by the dense foliage in which they roost, by their cryptic colouration (hair patterns and wrapped wings make them resemble dead leaves), and perhaps by heterothermy (identified in small Nyctimene and Paranyctimene by Bartholomew et al., 1970), which, although it may inhibit rapid escape, may reduce odour.
For some taxa there can be dramatic seasonal changes in roost composition. Most colonies of Eidolon helvum helvum use the same roosts for many years, but because of local fluctuations in food availability, some colonies make regular seasonal migrations, returning after a few months to their former roosting sites (Happold, 1987). In Pteropus poliocephalus colonies in Australia, the largest numbers are present in early summer when food is plentiful. In the southern part of the range, copulation takes place at the end of March or April, after which the camps break up and disperse because of scarcity of food.
Regular use by fruit bats can result in the denuding of the main branches used for roosting. (Photo by P. A. Morris)
Fruit bats feed almost exclusively on plants, taking floral resources (largely nectar and pollen but also petals and bracts), fruit (i.e. any plant material surrounding seeds), and often the seeds themselves and leaves (Marshall, 1985). Specialist seedeaters have not evolved as they have in birds (Snow, 1971). Insect remains have been found in the alimentary canal or intestine of megachiropteran bats (e.g. Lim, 1973; Start and Marshall, 1976) but their ingestion is perhaps accidental. However, Roberts and Seabrook (1989) observed Pteropus seychellensis aldabrensis on Aldabra Atoll feeding on ‘honeydew’ exuded by coccoids (Icerya seychellarum) present on a fig tree (Moraceae:Ficus lutea). They considered this to be an important food source and the bats to be an important control on numbers of coccoids. In the Sikkim province of India, Rousettus (Rousettus) leschenaulti has been recorded as feeding on fish (S. Mistry, pers. comm.). Fruit bats may also require extra water and have been observed drinking, sometimes taking seawater (Kock, 1972; Kingdon, 1974; Bergmans, 1978a).
Certain genera have remarkably catholic feeding habits. Marshall (1983) recorded that Eidolon helvum fed on flowers of 10 genera, fruit of 34 and leaves of 4. Similarly, the genus Pteropus used flowers of 26 genera, fruit of 62 and leaves of 3 (Marshall, 1983). Racey and Nicoll (1984) recorded Pteropus seychellensis seychellensis as feeding on 27 plant species from 14 genera (Table 1). In some cases, food preference and availability varies with place and season, but there are a number of indications that fruit bats may show food preferences if choice is available (Start, 1974; Marshall, 1985). For example, in West Africa, Eidolon helvum appears to favour Ceiba flowers (Bombacaceae) to Parkia flowers (Leguminosae), and Chlorophora fruit (Moraceae) to Solanum fruit (Solanaceae). As our knowledge increases, most megachiropteran bats will probably be seen not to be true generalists but rather ‘sequential specialists’, favouring at any one time and place one or a few plant species amongst the group of potential food plants available at that season (Marshall, 1985).
Plant genera are visited by a wide variety of bats. Thus the flowers of Ceiba attract at least 11 genera of Megachiroptera, and the fruit of Ficus at least 13 genera. Bats will feed upon both flowers and fruit of certain genera such as Musa (Musaceae). The generally catholic nature of the bat/plant relationship is supported by the fact that New World plants attract Megachiropteran bats, and Old World bat plants attract New World phyllostomids (e.g. flowers of Durio zibethinus (Bombacaceae) and Musa species) (Gardner, 1977).
Table 1. Food plants of Pteropus seychellensis seychellensis (after Racey and Nicoll, 1984).
A few flower or fruit species are largely associated with a single bat species. Thus Gould (1978) has shown that Oroxylum flowers (Bignoniaceae) are morphologically adapted for pollination by Eonycteris spelaea, although both Cynopterus and Rousettus have also been reported as visitors to these flowers (McCann, 1940).
The actual quantities of food consumed each night are difficult to ascertain. For some African frugivores it seems likely that they ingest about their own weight of fruit each night (Jacobsen and Du Plessis, 1976; Marshall and McWilliam, 1982; Wolton et al., 1982). For the nectarivorous Macroglossus sobrinus (18–26 g) Start (1974) estimated that one individual required the nectar produced by two inflorescences of Musa malaccensis each night. As each inflorescence produced at least 1.8 ml of nectar this means at least 3.6 ml of nectar was consumed per bat per night.
The distribution of bats is largely dependent on the spatial and temporal variation of their food resources. In equatorial regions food may be available within a small area throughout the year, whereas in more seasonal regions food may be relatively scarce for months. Some bat species may roost singly close to their food, whereas others may roost in great colonies (camps) from which they must fly long distances to feed. For example, McWilliam (1985–86) found that the feeding behaviour of three highly colonial species of Pteropus in Australia was dominated by the establishment and subsequent defence of long-term feeding territories. Investigation using radio-tagging showed that such visits to feeding sites spanned at least 29, 37 and 23 consecutive days for P. poliocephalus, P. alecto and P. scapulatus respectively. Mean straight line distances between roost sites and feeding locations for the same species were 10, 21 and 29 km respectively. Refuging species frequently forage in flocks and may travel considerable distances to feed. Flock foraging is particularly effective for the exploitation of a rich ephemeral and widely spaced food source. Such foraging has been observed in a number of bat species. Mixed species flocks (Epomophorus gambianus, Micropteropus pusillus and Nanonycteris veldkampii) have been observed around bat flowers in West Africa (Baker and Harris, 1957, 1959; Marshall and McWilliam, 1982) and single species flocks of Eonycteris spelaea have been seen in Malaysia, and of Rousettus (Rousettus) aegyptiacus in East Africa (Start, 1972; Start and Marshall, 1974).
The food of megachiropteran bats tends to be conspicuous, often clumped, and generally abundant and easily harvested within the clumps. Interspecific competition may be limited by spatial and temporal separation. Thomas (1982) studied a savannah community in the Ivory Coast, West Africa. Analysis of the diets of the resident species (Epomops buettikoferi, Hypsignathus monstrosus, Rousettus (Lissonycteris) angolensis and Micropteropus pusillus) showed they selected fruits from three mutually exclusive foraging zones and had little dietary overlap, with one exception. Each of these zones was associated with a particular forest height and the habitat (rather than the food resource) was partitioned. Analyses of the feeding behaviour, population sizes and fruit biomass available in the habitat, suggested that the two species with the highest diet overlap (E. buettikoferi and M. pusillus) could coexist only because fruit biomass was superabundant through most of the year. Migrant species (Eidolon helvum, Myonycteris torquata and Nanonycteris veldkampii) passed through the community at the onset of the rainy season. When they were part of the community, E. helvum shared the canopy foraging zone with H. monstrosus and M. torquata shared its foraging zone with R. angolensis. In the first case there was low overlap in diet species since E. helvum selected smaller fruits. In the second case there was more overlap in the diet, but coexistence appeared to be possible because R. angolensis could not fully exploit the peak in fruit productivity associated with the wet season.
The flowers of Sonneratia alba, an important mangrove timber tree, are especially ‘designed’ to attract bats for pollination. (Photo by W. E. Rainey)
Populations of four genera, Eidolon, Epomophorus, Pteropus and Rousettus, undertake seasonal migrations in those parts of their ranges where there are distinctive wet and dry seasons. For example, Epomophorus wahlbergi is not migratory in Kenya but may be so in South Africa (Allen, 1939; Wickler and Seibt, 1976). Eidolon helvum is present throughout the year in the tropical forest zone of Africa, but its colonies apparently vary greatly in size with season. Thus in Kampala, Uganda, a large colony numbered about 250,000 bats from September–October through the wet season to the breeding season in April. The colony dispersed in the dry period of July–August, many bats forming small, scattered roosts within 80 km of Kampala, but others presumably moving great distances north or south of the equator (Mutere, 1966, 1980). This species was entirely unknown at El Obeid, over 1400 km north of Kampala in the semi-arid central Sudan, until neem trees (Meliaceae: Azadirachta indica) were planted there, a tree whose fruit is favoured by many bats (Ayensu, 1974).
Certain plants play a major role in bat nutrition. The most obvious are the figs (Ficus spp.), a genus of the greatest importance to frugivorous animals throughout the world. One critical feature of the biology of certain figs is the unusual fruiting phenology, fruiting occurring asynchronously, and each tree fruiting every 6–12 months (Medway, 1972). Most other bat plants are more synchronous and more seasonal in their production of fruit, and we may expect to find a sequential series of flowering and fruiting within a plant assemblage that supports a megachiropteran bat community (Marshall, 1985).
Fruit bats, particularly on islands, have few natural predators. A variety of birds of prey, both Falconiformes and Strigiformes, various reptiles including snakes and large lizards, and some carnivorous mammals prey upon them (Nelson, 1965a; McClure et al., 1967; Kingdon, 1974; Wolf, 1984; Heideman et al., 1987; White et al., 1988; Pierson and Rainey, 1992). Although predators may influence both feeding and roosting behaviour, they seldom cause serious loss to populations (Marshall, 1983). On the islands of Guam in the Pacific and Christmas Island in the Indian Ocean, introduced arboreal snakes have had, or are likely to have, a devastating effect on the resident bat populations. On Guam, the brown tree snake (Boiga irregularis) attacks juvenile fruit bats. Observations of bat colonies between 1984 and 1988 indicated virtually zero survival of juveniles beyond 1–2 months, because of snake predation (Wiles, 1987b). In recent years on Christmas Island, the colubrid snake Lycodon aulicus capucinus appears to have established itself, posing a serious future threat (Smith, 1988).
In the genus Pteropus, many species are island forms, either being confined to oceanic islands or, like P. hypomelanus, roosting only on islands but flying to the mainland to feed. Mobility must vary greatly from species to species. On isolated islands Pteropus must have all its food requirements met by the plants of that island, although food sources will vary with season (Baker and Baker, 1936; van der Pijl, 1956; Perez, 1973; Wodzicki and Felten, 1975; Cheke and Dahl, 1981). On mainland areas Pteropus is certainly highly mobile and may be nomadic rather than migratory. Of the four Australian species, P. conspicillatus does not migrate, the coastal P. poliocephalus undertakes local seasonal movements up and down the coast, P. alecto undertakes more restricted seasonal movements than P. poliocephalus and only the inland P. scapulatus undertakes major movements, although these are of an erratic nature, largely following the flowering of Eucalyptus (Myrtaceae) (Ratcliffe, 1932; Nelson, 1965b).
Bats are very important pollinators and seed dispersers in tropical forests throughout the world ((Marshall, 1983, 1985; Fleming et al., 1987; Fleming, 1988; Cox et al., 1991, 1992; Pierson and Rainey, 1992) and have shared a long evolutionary history with angiosperms. Angiosperms possibly arose in the South East Asian region, around 130 million years ago, and achieved worldwide dominance over the gymnosperms about 90 million years ago. The first formations that we might recognize as tropical rain forests date from perhaps 60 million years ago. Megachiropteran bats have been in existence for at least 35 million years. Frugivory amongst the Megachiroptera arose before nectarivory (Marshall, 1983).
The visits by Megachiroptera to flowers for food may result in the pollination of those flowers. This is known to be the case for 31 genera in 14 families, with members of the Bignoniaceae (8 genera) and Bombacaceae (6 genera) being particularly prominent (Marshall, 1985). Many so-called ‘bat flowers’ are notably well-adapted for bat pollination (Faegri and van der Pijl, 1979), but other animals may also be significant pollinators: for example, Banksia (Proteaceae) is pollinated by Pteropus but also by a rodent (Rattus sp.) and a marsupial mouse (Antechinus sp.) (Recher, 1981).
Megachiropteran bats feed upon at least 145 genera of fruit in 30 families of plants widely distributed throughout the angiosperms (Marshall, 1985). The most important families are the Palmae (16 genera), Anacardiaceae (10 genera) and Sapotaceae (8 genera). Generally, fruits are consumed when ripe, but this is not always so; for example Cocos (Palmae) fruits are eaten when small and immature. Large fruits, such as mango (Anacardiaceae:Mangifera indica), must be consumed in situ, but smaller fruits may be carried away from the parent tree before being devoured and the seeds ejected through the mouth or anus. The distance a seed is carried will depend on its size and the size of the bat: tiny seeds which pass through the alimentary canal of a large bat will be carried furthest. Cynopterus brachyotis (30 g) can carry a fruit of up to 75 g, but it will seldom carry it more than 200 m (van der Pijl, 1957). On the other hand, Pteropus vampyrus (800 g) can carry fruits over 200 g (van der Pijl, 1957; Marshall and McWilliam, 1982). Pteropus vampyrus can travel about 50 km each night to feed so that long-distance dispersal may sometimes occur. Many fruits eaten by bats are also favoured by other animals, in particular man.
Macroglossus sobrinus, a South East Asian blossom bat, helps pollinate the flowers of a wild banana plant (Musa sp.) (Photo by K. G. Heller).
In the Philippines, an increased germination rate was recorded for fig seeds (Ficus chrysolepis) taken from bat faecal masses. It is suggested that this was due to differential ingestion of viable over non-viable seeds (parasitized by wasps of the Family Agaonidae). Because long distance seed dispersal by fruit bats is primarily through faecal deposition, this makes bat dispersal much more effective than previously suggested (Utzurrum and Heideman, 1991)
On many oceanic islands, with their limited faunas, fruit bats are the only animals capable of carrying large-seeded fruits. In such ecosystems, fruit bats can be the single most important pollinators and seed dispersers. In island ecosystems in the south-west Pacific, fruit bats are considered to be ‘keystone species’, because significant declines in forest regeneration rates and diversity would accompany their extinction (Cox et al., 1991, 1992). Many Pacific plant species are assumed to be exclusively dependent on fruit bats for successful pollination (Marshall, 1983; Cox, 1984a; Marshall, 1985; Elmqvist et al., in press). P. Cox (pers. comm.) has estimated that at least 30% of forest trees on Samoa are bat-dependent. In Samoa during the dry season, 80–100% of the seeds deposited on the ground (seed rain) in lowland forest are transported by fruit bats (Cox et al., 1992). The role of fruit bats in more complex ecosystems has been the subject of limited attention (Thomas, 1983).
Many of the plants that benefit from pollination or seed dispersal by bats are economically important to man (Fujita and Tuttle, 1991; Wiles and Fujita, 1992). At least 443 products useful to man derive from 163 plant species that rely to some degree on bats for pollination or seed dispersal (Fujita and Tuttle, 1991). These products include timber, fruits, fibres and tannins that contribute significantly to world markets as well as less well known products, such as medicines and food items important in local economies. The increasingly popular durian fruit (Durio zibethinus) depends on bats for pollination, as does petai (Parkia speciosa and P. javanica) whose seeds are a popular food item in South East Asia. Fujita and Tuttle (1991) estimate the monetary value of these and a third product (the fruit of duku - Meliaceae:Lansium domesticum) to exceed $US4 million annually in Indonesia. Annual sales of petai are estimated to exceed $US1 million in Peninsular Malaysia alone (Ng, 1980). Twelve tree species dependent on bats for dispersal are major timber species in Malaysia, one of the largest timber exporters in the world. The kapok or silk-cotton tree (Ceiba pentandra), the fibre, bark and seeds of which are economically important, is pollinated by a large number of bat species in Africa and South America (Baker and Harris, 1959; Toledo, 1977) but pollinated solely by Pteropus tonganus in Samoa (Elmqvist et al., in press).
The reproductive biology of fruit bats is reviewed by Pierson and Rainey (1992). Fruit bats are long-lived animals with low reproductive rates. In general, females do not give birth for the first time until they are one or two years old (Asdell, 1964; Nelson, 1965a, 1965b; Thomas and Marshall, 1984), although some small species (e.g. Macroglossus minimus and Eonycteris spelaea) may give birth before they are one year old (P. D. Heideman, pers. comm.). Females generally give birth to one young at a time after a 4–6 month gestation (Marshall, 1947; Neuweiler, 1969; Racey, 1973). In Pteropus, although the young may fly at three months they usually are not weaned until they are 4–6 months old, and may remain dependent on their mothers for a year. Lifespan in the wild is not well-documented. Heideman and Heaney (1989) undertook a capture-mark-recapture study of fruit bats on Negros in the Philippines. For three species (Cynopterus brachyotis, Haplonycteris fischeri and Ptenochirus jagorii), animals marked as yearlings were recaptured at least three years following first marking and from this data minimum longevities were estimated to be four to five years. In contrast, captive Pteropus giganteus have lived over 30 years (Nowak, 1991). Minimum mortality rates for juvenile H. fischeri were estimated at 10–30% during the first two-thirds of lactation and combined sub-adult and adult survival at 60% to 80%.
All wild populations of Pteropus that have been studied, except those of P. mariannus yapensis (Falanruw, 1988a), P. mariannus mariannus (Wiles, 1987b) and P. pumilus (Heideman, 1987) have a well-defined breeding season, with one young per adult female per year (Pierson and Rainey, 1992). In some species, there is a period of delayed implantation, as shown in Eidolon helvum where copulation occurs in June or July but gestation does not begin until November, with the young born in mid- to late-March (Mutere, 1967; Fayenuwo and Halstead, 1974). In Haplonycteris fischeri there is an 8-month post-implantation delay in embryonic development (Heideman, 1988). Some other pteropodid genera follow the same general pattern as Pteropus (Mutere, 1967; Dwyer, 1975; Heideman, 1987), but most breed aseasonally or exhibit two birth peaks a year (Thomas and Marshall, 1984; Heideman, 1987). Females of these pteropodid species often copulate in a post-partum oestrus and, as a result, are capable of producing two young per year (Thomas and Marshall, 1984; Falanruw, 1988a; Makin, 1990).
The population biology of fruit bats is remarkably similar to that of primates and their limited reproductive capacity makes them especially vulnerable to catastrophic events, such as cyclones and typhoons, and unnatural predation such as overhunting (Pierson and Rainey, 1992). When subjected to drastic declines, bat populations take several years to recover. In particular, island taxa, with restricted distributions, are at risk of extinction (MacArthur and Wilson, 1967).
Available data on the status of many fruit bats suggest serious population declines throughout the range, due principally to habitat loss, overhunting, and, on islands, tropical storms (Pierson and Rainey, 1992).
Reports by early explorers and scientists suggested that densities of fruit bats were once high throughout the Old World tropics. For example, Peale of the US Exploring Expedition in the 1840s described the forest of Samoa as being infused with the odour of bats (Cassin, 1858). Today there is no odour of bats in the forest, and it is possible to visit Samoa and never see a fruit bat (E. D. Pierson, pers. comm.). In the Philippines, roosts of up to 150,000 bats (Pteropus vampyrus and Acerodon jubatus) were common as late as the 1920s, but the largest colonies now number no more than a few hundred individuals (Heaney and Heideman, 1987; Diamond, 1988). In Australia in 1930, Ratcliffe (1932) reported Pteropus‘camps’ of up to 10 km long and 1.3 km wide, with estimated numbers of up to 30 million. Now, many colonies have disappeared entirely, and only a few are reported to contain more than 100,000 individuals (Pierson, 1984). In a recently completed survey of bat utilization in Indonesia and Malaysia, Fujita (1988) reported that for the past 10 years hunters have been finding it increasingly difficult to locate Pteropus roosts.
Man's activities are the most important threat to fruit bats. Many species are dependent on primary forest and thus threatened by the large-scale destruction of rain forest in many tropical areas. Disturbance of roosts may be incidental or deliberate. Many fruit bats are hunted both at a local or commercial level. Commercial hunting of species in the Pacific area to satisfy the demands of consumers on the island of Guam has resulted in the decline of many populations and the extinction of at least one species. In some areas there has been conflict between bats and commercial fruit growers.
Bats are also threatened by natural factors. Tropical storms are an ever present hazard and can have devastating effects, particularly where populations are already under pressure from human activities. Disease is a factor whose importance is not yet fully understood.
Habitat loss has been cited by a number of authors as a major factor contributing to declines in fruit bat populations (Wodzicki and Felten, 1975; Racey, 1979; Cheke and Dahl, 1981; Carroll, 1984; Pernetta and Hill, 1984; Diamond, 1988; Fujita and Tuttle, 1991; Pierson and Rainey, 1992). Although information on habitat requirements is limited for some species, it is evident that there is considerable ecological variation within the family. Some species, like Pteropus gilliardi on New Britain and P. livingstonii on the Comoros, for example, appear to be confined to montane forests; others, like Pteropus conspicillatus in Australia or P. tonganus, frequent agricultural areas.
Deforestation, widespread in almost all tropical areas of the world, has had several identifiable consequences for fruit bat populations (US Fish and Wildlife Service and National Environmental Protection Board, 1989). Many species, particularly those inhabiting mangrove swamps (e.g. Pteropus vampyrus in Malaysia and Indonesia) and lowland forest, have lost critical roosting areas. Mangrove swamps are being destroyed by the woodchipping industry, for mariculture, firewood, and coastal development, and lowland forest is felled for agriculture and timber.
Loss of forest results in the loss of critical food resources for many species. The loss of tamarind trees (Leguminosae: Tamarindus indica), a favourite food of Pteropus rodricensis, has been identified as one factor in the decline of this species (Cheke and Dahl, 1981). Even Pteropus tonganus, which appears adaptable to agricultural conversion, greatly preferred native to cultivated fruits in a recent feeding trial (E. D. Pierson, pers. comm.).
Urbanization involves road building and easier access to remote roosting areas (Falanruw, 1988a). This means it has been easier to hunt animals at their roosts. Such disturbance can cause animals to abandon roost sites (Wiles, 1987b), with serious consequences, particularly during the maternity season.
On many islands forest loss due to human depredation is exacerbated by tropical storms, because remnant forest is particularly prone to damage by high winds.
Many fruit bat species are strongly colonial and this makes populations vulnerable to disturbance at their roost sites. In the case of cave-dwelling fruit bats, populations are threatened by over-exploitation for bat guano (originating mainly from the insectivorous bats with whom the frugivores share these caves), by mining and quarrying of the caves themselves or of the adjacent environment, and other uses to which caves are put (such as religious worship, tourist attractions, or even as human habitation). Examples can be found where visitors to temple caves are not a major problem to the bats (such as in Bali), but this has proved a problem in others. Improved education about bats can also result in conservation problems: tourists now want to see spectacular bat colonies and there is already evidence of unacceptable levels of disturbance caused to cave-dwelling bats by uncontrolled visits to well-known bat caves. In Thailand there have been reports of deliberate disturbance of cave-roosting bats by tour guides (Hutson, 1990). Similarly, the colony of Pteropus vampyrus in the Botanical Garden at Bogor in Indonesia is protected but nevertheless often disturbed (and sometimes even hunted) by the guards, to please the tourists (W. Bergmans, pers. comm.). It should be stressed, however, that careful use of colonies as educational tools has proved very successful, as shown in Australia with the Gordon fruit bat colony in Sydney.
Logging is a special problem for the many fruit bat species that are restricted to small islands. (Photo by W. E. Rainey)
Interactions between bats and commercial fruit: In some areas of the world (for example, Australia, Israel and South Africa) large-scale commercial fruit growing has led to conflicts between fruit growers and bats (Jacobsen and Du Plessis, 1976; Loebel and Sanewski, 1987; Makin and Mendelssohn, 1987). Many cultivars have been developed from wild species that are dependent upon bats for pollination or seed dispersal, or both (van der Pijl, 1957; Marshall, 1983). The same characteristics (colour, smell, taste) that attract bats to wild species may also attract them to cultivated ones, although in the latter case they can rarely be beneficial as pollinators or seed-dispersers as this role has largely been supplanted by the fruit grower. One exception to this is the durian, which still relies heavily on bats for its pollination. The most serious conflicts may occur where the supply of native fruits has been reduced through forest loss (Fleming and Robinson, 1987; Tidemann and Nelson, 1987) or where there has been a mass failure of native plants to flower, as has happened with Eucalyptus in Australia, whose blossoms provide the predominant food for Pteropus species there (Ratcliffe, 1931; Nelson, 1965a).
In most cases bats feed on fruit that is too ripe to be marketable. Many fruits are picked when they are unripe and allowed to ripen off the tree. The ripening of fruit is mediated by the action of ethylene (Burg and Burg, 1965), which is produced naturally in the plant, and which is produced in increasing quantities as ripening progresses (Burg, 1962). Any trauma to the plant tissue (e.g. bites or scratches) can also lead to increased ethylene production (Yang, 1981) and in some cases premature ripening. Thus damage can be caused indirectly by bats clambering over unripe fruit or through ‘test bites’ (Ratcliffe, 1931; van der Pijl, 1957). The relative perceived scale of damage has also increased through increased market demands for unblemished fruit.
The level of damage varies considerably with locality and is generally greatest in the summer when females are lactating and have greater energy requirements.
Crop protection and management of fruit bats: In the 1930s in Australia, Ratcliffe (1931) concluded that the impact of fruit bats on fruit-growing was insignificant. As available natural habitat declined and the fruit-growing industry expanded, so the problem became economically more important (Tidemann and Nelson, 1987) and resulted in protective legislation for fruit bats being repealed in 1984 in Queensland.
Destructive management methods such as shooting have been favoured in the past and persist today. However, a growing body of opinion favours non-destructive methods, and a thorough investigation of the effectiveness of such methods is a high priority.
The methods of protection and control can be grouped into three main categories: management at roosts, protection at orchards and management of farms.
1. Fruit bat roost management.
Dispersing daytime roosts is still the most common approach to management. Shooting is the most popular method though other forms of harassment have included the use of helicopters, bird-scare guns, dense smoke and loud noise. In these cases the colony may simply end up moving to a nearby location (Palmer, 1987).
More drastic methods of management include the total removal of trees and draining of swamps. Again, this may not necessarily move bats from the general area. Fruit bats can travel 40 to 50 km in a night to reach feeding areas, as has been shown for Eonycteris spelaea (Start, 1974), and have a well-developed memory of the landscape they utilize (G. C. Richards, pers. comm.).
Bus advertising control of fruit bats, Harare, Zimbabwe. (Photo by A. M. Hutson)
In Israel, campaigns were undertaken against Rousettus (Rousettus) aegyptiacus by fumigation of their cave roosts. As well as killing the target species it also resulted in heavy losses amongst insectivorous bats that shared the cave roosts (Makin and Mendelssohn, 1987).
In the Maldives, drastic control measures have been instigated that could threaten the resident bats, Pteropus giganteus ariel, and P. hypomelanus maris, with extinction (see relevant species accounts). Accurate estimates of crop damage and population numbers were not made before netting started, reducing some island populations by up to 80% (Dolbeer et al., 1988).
If considered necessary, roost management can be carried out on a sustainable basis, as has been shown by Halstead (1977), who detailed the management programme at the University of Ife, Nigeria. There, controlled culling of the resident colony successfully provided income through the sale of bat meat, and animals that were used for teaching and research.
In general destructive roost management methods are at best ineffective and at worst highly damaging to target and nontarget bat species.
2. Crop protection in orchards
Many techniques and devices have been used in attempts to protect orchards against fruit bat raids.
A number of techniques have met with initial success, but proved unsuccessful in the longer term. Scare guns probably deter bats from making an initial visit but could end up attracting bats. Similarly the efficacy of ultrasonic scarecrows is doubtful (Fleming and Robinson, 1987). Other sonic devices have been tried but their use is limited (Calford and McAnally, 1987). Flashing strobe lights and bright light grids over orchards have been initially successful, but it appears that bats become accustomed to the lights and will feed in a fully illuminated orchard as they will in suburban fruit trees. Bright lights may also attract bats, sometimes acting as a beacon to guide them to the orchard each evening. Harassment by random shooting, beating metal drums and smokey fires have all been used with some success, but depend upon regular use throughout the night. The unpredictable nature of fruit bat raids and the long man-hours involved make these techniques unattractive for fruit-growers.
Other techniques have met with more success. Replaying recorded sounds of bat distress calls has shown some promising results, but further studies are required. The smell of carbide has been claimed to be successful in deterring bats from litchi trees (Sapindaceae: Litchi chinensis) in northern Queensland (Watson, 1982). Fruit bats have a highly developed sense of smell and observations on captive bats (H. Luckhoff, pers. obs.) show they have an extreme aversion to the smell of fresh meat, particularly liver. Tests on a bird-repellent, Methiocarb, indicate that this may be suitable for fruit bats (M. Tuttle, pers. obs.). It is a short-lived carbonate, which breaks down in sunlight and is a powerful emetic. Birds soon associate its effects with the fruit they eat. No long-term effects, or deaths, have occurred during trials. Netting is the most effective method of protecting trees from fruit bats (Loebel and Sanewski, 1987). It has the added advantage of keeping out other pests such as birds and possums, although it is expensive and needs regular maintenance. Similarly, electric wires may be effective for small orchards. To be effective for fruit bats the wires must be no more than 25–30 cm apart with additional dangling wires forming a circular curtain around each tree (Anon., 1983).
Out of all of these techniques, netting is the most effective, though it is expensive to erect and maintain.
3. Farm management
The careful management and siting of fruit farms may yield the most effective results.
Decoy trees are showing promising results in several areas where rows or buffer zones of other fruit trees such as native figs are used. More research is required on the species of tree used, the time they take to produce fruit and their season of fruiting in relation to the commercial crop.
Early picking of the fruit crop is an effective management strategy (Tuttle, 1985) but may not be suitable for all fruit. Bats may damage quite small and green fruits (e.g. peaches [Rosaceae: Prunus persica]) if natural food sources are scarce, and developing fruit when visiting blossoms for nectar or when visiting early ripened fruit (e.g. bananas [Musa spp.]).
The removal of early ripened and over-ripe fruit left after picking should be considered in farm management. Both these situations are known to attract fruit bats and in the latter case, may attract fruit bats to nearby orchards, which may not yet have picked their fruit.
Damaged fruit can be used for juicing and canning, but, depending on the crop, cuts the fruit-growers' profits considerably. Fruits such as mangoes, still intact but rendered unmarketable because of teeth marks, are one example where income could be gained by alternative uses.
Growing alternative crops and shifting the locality of the farm may be the best solution in areas where fruit bats are a problem. New fruit farms should be sited with care. There are still regular examples of new orchards being established in close proximity to large permanent fruit bat colonies, even where natural food supplies are reduced by the removal of native forests. Crops that need to ripen on the tree should be avoided in known fruit bat areas. If seasonal movements of fruit bats are known to occur in an area, fruit species that will crop while fruit bats are absent would be more suitable to grow. Orchards should be laid out in blocks or segments of a size that facilitates techniques for crop protection, such as netting. In conclusion, there is no cheap or simple answer to the problem of fruit bat damage to crops. The major cause of the problem is related to loss of natural habitat coupled with a growth in the fruit-growing industry. Destructive management techniques should be banned and research into the effectiveness of other non-destructive methods stepped up. Above all, no management scheme should be instigated without first assessing the level of damage and the likely long-term threat to the target species. This can be achieved only through closer cooperation between fruit growers and conservationists.
Bats as a food item: In many areas bats have for a long time been important as food for local people. There are records of fruit bats as a major dietary item from most of their range, from Guam (wiles, 1987a, 1987b), Vanuatu (Chambers and Esrom, 1991), Samoa (Cox, 1983), the Cook Islands (Wodzicki and Felten, 1980), the Philippines (Heaney and Heideman, 1987), the Togian Islands (Owen et al., 1987; Hill, in press), Irian Jaya (Craven, 1988), Thailand (Lekagul and McNeely, 1977), Indonesia and Malaysia (Fujita, 1988) and the Seychelles (Racey, 1979; Cheke and Dahl, 1981). In South East Asia, fruit bat meat is also valued as a remedy for asthma, kidney ailments, and ‘tiredness’, especially among people of Chinese origin (Fujita and Tuttle, 1991). In Nigeria, bats are occasionally found in ju-ju stalls because they are thought to cure barrenness (and to promote fertility) in women (Shoga, 1974).
In some areas, such as in the Marianas in the western Pacific, fruit bats are considered a delicacy and are served at social occasions such as village fiestas, weddings, christenings and holiday celebrations (Wiles and Payne, 1986). In many local markets, bats have a considerable commercial value (Wiles and Payne, 1986). In north Sulawesi, up to 16 species of bats were found to be available in local markets, including some species hitherto considered rare (Bergmans and Rozendaal, 1988). Fruit bats are a luxury item on restaurant menus in many parts of their range (Anon., 1988b, 1988c; Fujita and Tuttle, 1991). In the Seychelles, restaurants advertise bat curry, and Racey (1979) estimated that one restaurant could use up to 1500 bats a year.
Hunting and the decline in bat populations: Traditionally, fruit bats have been hunted using methods such as thorny vines, nets or fish hooks (Cox, 1983; Fujita and Tuttle, 1991). The introduction of firearms and the transition from subsistence to commercial harvesting has resulted in declines in bat populations in many parts of their range (Wodzicki and Felten, 1975; Engbring, 1985; Wiles, 1987b). In 1988, several Samoan chiefs indicated that declines in local bat populations began with the introduction of guns (P. Cox, T. Elmqvist, E. D. Pierson and W. E. Rainey, pers. comm.). Overhunting has been depleting fruit bat populations since 1930 (Wiles and Payne, 1986), and any cultural limitation on the exploitation of this resource has long since been forgotten (Falanruw, 1988a).
A flying fox (Pteropus tonganus) packed for sale in a Saipan supermarket. (Photo by G. J. Wiles)
The level of hunting in Malaysia and Indonesia is difficult to estimate but a survey in 1985–86 revealed that hunting for human consumption was quite common (Fujita and Tuttle, 1991). Fujita and Tuttle (1991) reported that figures provided by many bat vendors suggest that the annual sales of a single vendor could be about 10,000 bats a year, enough to eliminate an average sized colony each year.
In northern Thailand, R. E. Stebbings (pers. comm.) reported that hunters from large towns made trips to known bat caves in order to net bats (both fruit-eating and insectivorous) with serious effects on populations.
In Africa the hunting of bats is known to occur in Guinea (J. R. Wilson, pers. comm.) and Nigeria (Happold, 1987). The degree to which hunting affects bat populations, whether in these two sites or elsewhere, is unknown.
In addition, the most intense hunting for local use seems to occur during the bats' reproductive season. Hunters in Indonesia, Malaysia, Samoa and the Cook Islands (Wodzicki and Felten, 1980; Fujita and Tuttle, 1991; La Mositele, pers. comm.) identify a ‘bat season’, which coincides with the main fruiting or flowering peaks. During this time, females are often caught pregnant or with attached young. This seasonality of hunting has important implications for the ability of populations to recover from intense hunting pressure.
Guam and the trade in the Pacific: A comprehensive overview of the history of the Pacific bat trade is given in Wiles (1992).
By far the most serious threat to bat populations has come from the trade in fruit bats centred on Guam. Archaeological evidence indicates that the Chamorro people of Guam and the other Mariana Islands have been eating fruit bats for over 1000 years (Lemke, 1986). While traditional hunting methods were used, the harvest had little impact on bat populations (Lemke, 1986). However, with the introduction of firearms and a cash economy populations declined, with the endemic species, Pteropus tokudae, becoming extinct. Because of these declines, Guam's residents began importing fruit bats from elsewhere in the Pacific as early as the late 1960s (Wiles and Payne, 1986). Statistics on annual imports of fruit bats have been compiled by the Guam Division of Aquatic and Wildlife Resources since 1975 (Table 2). Between 1975 and 1989, a total of 220,899 fruit bats were imported into Guam.
Table 2. Summary of numbers of fruit bats imported into Guam in the period Financial Years 1975–1989. Data are largely lacking for 1987. After Wiles, 1992.
The main species involved in trade is Pteropus mariannus. However, other species have been seen in trade. Pteropus vampyrus, P. hypomelanus and Acerodon jubatus have been exported from the Philippines, and P. samoensis from Western Samoa. In the last case, this species is the much less common of the two resident Western Samoan species (P. Cox, E. D. Pierson and W. E. Rainey, pers. comm.).
Many countries have supplied Guam with fruit bats (Table 2). The pattern of trade has shifted in response to changes in legislation and political status of the countries concerned. The political map of the western Pacific is complex (see Figure 2) and trade between countries and Guam (an unincorporated US Territory) has been classed as internal or international depending on the political status of the exporting country. This has had serious implications because CITES (Convention on International Trade in Endangered Species of Wild Fauna and Flora) covers only international trade. For example, the Commonwealth of the Northern Mariana Islands (CNMI) is a US Commonwealth, and trade between the CNMI and Guam is classed as internal and not covered by CITES, while the Federated States of Micronesia (FSM), formerly part of the US Pacific Trust Territory, is now independent and thus trade is covered by CITES. Changes in legislation have not necessarily curbed the trade in fruit bats. In the CNMI, despite a series of one- and two- year hunting moratoria first established in 1977, trade has continued, mostly from the southern, heavily populated islands of Saipan, Tinian and Rota. This has resulted in declines in numbers of resident bats. Exports from Pohnpei and Chuuk in the FSM have continued since independence in 1986, despite the trade being covered by CITES. However, in other countries, enactment of legislation has curbed trade, as in Yap (FSM, legislation enacted in 1981), American Samoa (legislation enacted in 1986) and Western Samoa (legislation enacted 1989). In 1989, the CITES regulations were amended so that all Pteropus and Acerodon species were on either Appendix I or II. Appendix I prohibits trade in all but a few exceptional circumstances, while Appendix II controls, but does not prohibit, international trade. During the first three months of 1990 (before enforcement) Guam imported an estimated 3989 fruit bats, while in May and June imports dropped to 292 (Wiles, 1990). However, there remains one significant loophole. Belau, a trust territory of the US, has continued to export fruit bats legally to Guam and Saipan because of this being internal trade. Belau has been the major supplier to the Guam market since 1975, exporting over 110,000 bats between 1975–89 (Table 2). A decline in shipments in mid-1990 was short-lived, and between October 1990 and February 1991, Belau sent about 3800 bats to Guam and continued to send bats to Saipan in the CNMI (G. Wiles, pers. comm.).
Figure 2. Political map of Pacific Islands. (Reproduced with kind permission of the Hawaii Geographic Society)
There is also a continuing illegal trade in fruit bats into Guam. In the period 1986–88 Guam Customs and Quarantine Division confiscated 1580 bats in 118 shipments (Wiles, 1986, 1987c, 1988). Bats are reported to be smuggled in coolers filled with fish from Rota to Guam (Lemke, 1986), and similarly from Western Samoa to American Samoa and then on to Guam.
Finally, while the trade has centred on Guam, there has been trade in other countries in the Pacific. M. S. Fujita (pers. comm.) reports that both Pteropus vampyrus and P. hypomelanus are traded extensively in South East Asia. Malaysian bats appear in markets in Singapore and probably Jawa; bats from Irian Jaya may be traded in Jawa. Vanuatu exported 365 Pteropus tonganus to Noumea, New Caledonia in 1989 and early 1990 (E. Banei, pers. comm.).
Fruit bat advert from Pacific Daily News. (Photo by G. J. Wiles)
Future problems and needs: Amendment of the CITES regulations has helped to curb the trade significantly, although the situation in Belau remains worrying. The enforcement of CITES regulations is critical to the survival of many endangered Pacific species. In the past the trade has shifted rapidly from country to country in response to changes in regulations and supply (i.e., depletion of stocks), and there are a number of countries (for example in South East Asia) that might enter into trade should other sources close down (G. Wiles, pers. comm.). Determining the source of some bats is made difficult by problems of identification and there is an urgent need for an identification key to those animals in trade.
In addition to the threats introduced by humans, bats on small islands are also vulnerable to natural pressures. The endemic Pteropus rodricensis was brought to the edge of extinction in Rodrigues by a combination of deforestation, hunting for food, and cyclones. In 1979 Cyclone Celine II reduced the number of animals from 151 to 70 (Jones, 1980; Carroll, 1984).
The effects of the typhoon that hit Western Samoa and American Samoa in February, 1990 on Pteropus tonganus and P. samoensis are still being evaluated. After the typhoon the bats were foraging for fruit on the ground or at fallen trees in villages, not always at night. Since they were often unable to take flight from the ground they were extremely vulnerable to predation. Domestic animals (dogs, cats and pigs) were reported to have killed large numbers of P. tonganus. Since cats and pigs also forage extensively in agroforest, the mortality was probably greater than that directly observed (E. D. Pierson pers. comm.).
For nearly a week in 1986, Malaita, an island in the southern end of the Solomon Islands, was pounded by Cyclone Namu, causing a significant loss of human life. During the cyclone's passage over the southern islands, considerable damage was done; landslides denuded large areas, and vast tracts of forest were stripped of leaves and fruit. Flannery (1989) visited Sinalaggu harbour, off Malaita's east coast in December 1987, and was shown hundreds of lower jaw bones of fruit bats that had been collected in the few months following the cyclone. The great majority of the bones were of Pteropus rayneri but a few were P. tonganus.
Evidence from the Marianas, Samoa and Vanuatu suggests, however, that a major cause of storm-related mortality of fruit bats is intense post-storm hunting (Pierson and Rainey, 1992). Defoliation reduces concealment of roosting animals, so they are more readily hunted. Because of decreased food availability, bats may forage diurnally and become less cautious, increasing the risk of opportunistic killing. Perhaps more important is the long-term damage to already damaged forests and the time that it may take for forests to recover, if they are indeed allowed to. The long-term survival of bats depends on the status of populations and the storm intensity. Forests on Rota (in the CNMI) recovered after being hit by a storm but the bat population dropped to a new low, further increasing the risk of extinction (E. D. Pierson, pers. comm.).
Forest re-sprouting after typhoon damage on Savai'i, Western Samoa. (Photo by W. E. Rainey)
What little is known about disease in fruit bat populations is summarized by Pierson and Rainey (1992). The first evidence of severe epidemics decimating wild populations comes from the Whitney Expeditions in the 1930s. In a 2-month survey of Kosrae in the Federated States of Micronesia, researchers located only four bats (Pteropus mariannus ualanus) and learned from local residents that the other animals had all died in a recent epidemic associated with an outbreak of measles in the human population (Coultas, 1931). Degener (1949) reports on a similar epidemic depleting P. tonganus populations near Savu Savu, Fiji, sometime prior to 1949. More recently, Flannery (1989) described epidemics in two bat populations. In June 1988 Flannery visited Manus, the largest of the Admiralty Islands, north of the New Guinean mainland. During a week of searching in central Manus, he failed to find P. neohibernicus, although the smaller P. admiralitatum was abundant. Local people reported that in 1985 many P. neohibernicus were found dead and dying, presumably from disease, under large and well-known roosts. The deaths occurred throughout the island over a period of a few weeks, and afterward no large fruit bats were seen for several years. Just before the time of Flannery's 1988 expedition, several hunters reported having seen an occasional large fruit bat, suggesting that the entire population had not been wiped out. Flannery (1989) reports on a similar incident on the islands of Bougainville and Buka in the northern Solomons in 1987. In this case, the dead bats were largely or entirely P. rayneri grandis. Populations of P. neohibernicus on the islands of New Ireland and New Britain, which lie between Manus and Bougainville, were examined in 1988 and 1989 and had suffered no such decline.
On Manus and Bougainville the high fatality rate and exceptional nature of the epidemics suggest that the organism responsible was not endemic, and indeed may have been newly introduced into these populations by domestic animals (Flannery, 1989).
Bats receive protection both at national and international levels. Full details of national legislation can be found in International Union for the Conservation of Nature and Natural Resources Environmental Law Centre (1986) (for Africa), and Nichols et al. (1991) (for Asia and Oceania). Details of international legislation can be found in Lyster (1985).
There is a great variation in how fruit bats are treated under national legislation and the sections below are arranged according to the level of protection, or otherwise, that bats receive.
Countries giving full protection: Only Ethiopia protects all members of the Family Pteropodidae. Other countries give full protection to certain species. Thus Rousettus (Rousettus) leschenaulti seminudus is fully protected in Sri Lanka, as is Pteropus niger on Réunion, P. rodricensis in Mauritius and P. mariannus mariannus on Guam.
Countries giving partial protection: Madagascar lists the resident Pteropus rufus as noxious, but charges a fee for holders of commercial permits to allow them to take bats. Malaysia (federal), under its Protection of Wildlife Act of 1972, partially protects and prohibits or regulates possession or national trade and international trade of Pteropus hypomelanus and P. vampyrus. New Caledonia gives partial protection to fruit bats by the regulation of hunting. In Papua New Guinea the use of mist nets to take bats is prohibited without prior permission of the conservator. In Burkina Faso there is partial protection for Epomophorus and Myonycteris species, although other bats are exempted from the wildlife regulations. In FijiPteropus tonganus and Notopteris macdonaldii are protected through regulation of international trade. In the Commonwealth of the Northern Mariana Islands (CNMI) 1- and 2- year moratoria on hunting have been established. In Yap there is a ban on the taking and exporting of bats. Similarly, in American Samoa no commercial harvest of bats is allowed and export and hunting are regulated. In Western Samoa, commercial export of bats is banned. In Belau the only protection given is that bats must be harvested using nets. Nepal gives partial protection to all fruit bats.
Countries exempting bats from wildlife regulations: Benin, Ivory Coast and Togo exempt bats from their wildlife regulations, as does Burkina Faso, although here two genera are given partial protection (see above). Pakistan exempts bats from the regulation of international trade, while the Punjab area of Pakistan specifically excludes Pteropus giganteus from protection. South Africa exempts fruit bats from its regulations. In Australia a number of species are not protected, although in some cases national and international trade is regulated. Thus, Pteropus alecto is unprotected in Queensland and Western Australia, P. poliocephalus is unprotected in Queensland, P. scapulatus is unprotected in Queensland, New South Wales and Western Australia and Macroglossus minimus is unprotected in Western Australia.
Countries listing bats as ‘noxious’: India, Indonesia and Israel specifically list fruit bats as ‘noxious’, although in the case of Indonesia this refers to ‘fruit-eating bats’. In the Northern Territory of Australia, Pteropus alecto and P. scapulatus are listed as ‘noxious’.
It can be seen that the national laws protecting bats are very varied and in some cases very complex. There are also some anomalies. For instance, Pteropus rodricensis is given full protection under United States regulations, and partial protection by Natal in South Africa. Similarly, Aproteles bulmerae (from New Guinea) is given full protection under United States regulations, but under the Lacey Act bats of the genus Pteropus are listed as injurious animals.
Full details of the structure of CITES are given by Lyster (1985).
In October 1989, CITES member states approved proposals to include seven species of fruit bats in CITES Appendix I (Table 3) and all six species of the genus Acerodon and the remaining Pteropus species in CITES Appendix II. The Appendix I listing provides for a prohibition on international trade in the most threatened species, while the Appendix II listing provides for regulation of international shipments containing other Pteropus and Acerodon species. This decision was a major achievement for conservationists and promises substantive international protection for these species for the first time.
Table 3. Pteropus species included in CITES Appendix I as of 18 January 1990.
Belau's status as the last remnant of the US Pacific Trust Territory has allowed for exports to Guam to bypass CITES requirements because the trade is viewed as internal under CITES. This continues to cause concern. The major challenge in the future will be the enforcement of the new CITES regulations. The US Fish and Wildlife Service is responsible for enforcement of CITES and it is encouraging that after a period of uncertainty money was allocated in the 1991–92 federal budget to fund the post of enforcement officer for one further year. The importance of making this post permanent cannot be over-stressed. While effective enforcement of CITES has obvious implications for the species transferred to Appendix I, there must be equally concerted action to monitor the trade in Appendix II species, which may become subject to trade pressure as a result of the Appendix I listing. Such action will be possible only through increased involvement of federal resources.
Confiscated dead fruit bats from American Samoa (Photo by G. J. Wiles)
The Protocol on Protected Areas and Wild Fauna and Flora of the Eastern African Region is open to the contracting parties to the above convention. It was adopted in 1985 and as of the beginning of 1992, it was not yet in force. The protocol provides for the protection of threatened and endangered species of flora and fauna and important natural habitats in the Eastern African region. Pteropus niger and P. rodricensis are covered under this protocol.
This convention was signed in 1968, and as of 1985 28 states were Parties to the Convention with a further 14 having signed but not ratified (Lyster, 1985). The Convention is primarily concerned with wildlife but also embraces the conservation of other natural resources such as soil and water. It emphasizes the need for protected areas and special conservation measures for species listed in an Annex. It also covers topics such as conservation education, research and the need to integrate conservation into development plans. Unfortunately, it has not established an administrative structure to oversee its enforcement and as a result little has been done to encourage Parties to implement its provisions (Lyster, 1985). No further progress on implementation had been made by the beginning of 1992.
This Convention was adopted in 1975 with an objective of protecting natural and cultural areas of ‘outstanding universal value’. In 1989, 315 sites were on the World Heritage List including some that are known to be or potentially could be of significance for bats. One such site is Aldabra Atoll in the Indian Ocean.
The Bonn Convention was concluded in 1975, although it did not come into force until 1983. As of the beginning of 1992, there were 39 Parties to the Convention. The Convention aims to protect migratory species through imposing regulations on range states that exercise jurisdiction over any part of the range of a migratory species. Migration is taken to include any regular cross-border movements. This Convention may have future relevance for fruit bats, particularly in Africa, where movements are regularly recorded.