Structure and diversity of phyllostomid bat assemblages on riparian corridors in a human‐dominated tropical landscape

Tropical forests around the world have been lost, mainly because of agricultural activities. Linear elements like riparian vegetation in fragmented tropical landscapes help maintain the native flora and fauna. Information about the role of riparian corridors as a reservoir of bat species, however, is scanty. We assessed the value of riparian corridors on the conservation of phyllostomid bat assemblage in an agricultural landscape of southern Mexico. For 2 years (2011–2013), mist‐netting at ground level was carried out twice during the dry season (December to May) and twice during the wet season (June to November) in different habitats: (1) riparian corridors in mature forest, (2) riparian corridors in pasture, (3) continuous forest away from riparian vegetation, and (4) open pastures. Each habitat was replicated three times. To determine the influence of vegetation structure on bat assemblages, all trees (≥10 cm dbh) were sampled in all habitats. Overall, 1752 individuals belonging to 28 species of Phyllostomidae were captured with Sternodermatinae being the most rich and abundant subfamily. Riparian corridors in mature forest and pastures had the greatest species richness and shared 65% of all species. Open pastures had the lowest richness and abundance of bats with no Phyllostominae species recorded. Six of the 18 species recorded could be considered as habitat indicators. There was a positive relationship between bat species composition and tree basal area. Our findings suggest that contrary to our expectations, bats with generalist habits and naturally abundant could be useful detector taxa of habitat modification, rather than bats strongly associated with undisturbed forest. Also in human‐dominated landscapes, the maintenance of habitat elements such as large trees in riparian corridors can serve as reservoirs for bat species, especially for those that are strongly associated with undisturbed forest.     

Heard but not seen: Comparing bat assemblages and study methods in a mosaic landscape in the Western Ghats of India

We used capture (mist‐netting) and acoustic methods to compare the species richness, abundance, and composition of a bat assemblage in different habitats in the Western Ghats of India. In the tropics, catching bats has been more commonly used as a survey method than acoustic recordings. In our study, acoustic methods based on recording echolocation calls detected greater bat activity and more species than mist‐netting. However, some species were detected more frequently or exclusively by capture. Ideally, the two methods should be used together to compensate for the biases in each. Using combined capture and acoustic data, we found that protected forests, forest fragments, and shade coffee plantations hosted similar and diverse species assemblages, although some species were recorded more frequently in protected forests. Tea plantations contained very few species from the overall bat assemblage. In riparian habitats, a strip of forested habitat on the river bank improved the habitat for bats compared to rivers with tea planted up to each bank. Our results show that shade coffee plantations are better bat habitat than tea plantations in biodiversity hotspots. However, if tea is to be the dominant land use, forest fragments and riparian corridors can improve the landscape considerably for bats. We encourage coffee growers to retain traditional plantations with mature native trees, rather than reverting to sun grown coffee or coffee shaded by a few species of timber trees.     

Ectomycorrhizal colonization of naturally regenerating <Emphasis Type="Italic">Pinus sylvestris</Emphasis> L. seedlings growing in different micro-habitats in boreal forest

We investigated the species richness and composition of ectomycorrhizal (EM) fungi colonizing Pinus sylvestris L. seedlings naturally regenerating in boreal forest, in three different microhabitats: on forest ground, on decaying stumps, and within moss layer on erratic boulders. We tested the hypothesis that habitat differences would affect the composition of the EM community of regenerating pine seedlings. In total, 16 EM species were detected, from which none occurred on seedlings growing in all three microhabitats. Piloderma croceum and Cenococcum geophilum were common for seedlings growing in forest ground and on boulders, while Tricholoma aestuans and Suillus luteus were shared between seedlings growing on forest ground and decaying stumps. EM species richness and composition were strikingly different between seedlings regenerating in different microhabitats. Results are discussed as a function of dispersal and niche differentiation of EM fungi.     

Bats in a Farming Landscape Benefit from Linear Remnants and Unimproved Pastures

Schemes designed to make farming landscapes less hostile to wildlife have been questioned because target taxa do not always respond in the expected manner. Microbats are often overlooked in this process, yet persist in agricultural landscapes and exert top-down control of crop pests. We investigated the relationship between microbats and measures commonly incorporated into agri-environment schemes, to derive management recommendations for their ongoing conservation. We used acoustic detectors to quantify bat species richness, activity, and feeding in 32 linear remnants and adjacent fields across an agricultural region of New South Wales, Australia. Nocturnal arthropods were simultaneously trapped using black-light traps. We recorded 91,969 bat calls, 17,277 of which could be attributed to one of the 13 taxa recorded, and 491 calls contained feeding buzzes. The linear remnants supported higher bat activity than the fields, but species richness and feeding activity did not significantly differ. We trapped a mean 87.6 g (±17.6 g SE) of arthropods per night, but found no differences in biomass between land uses. Wider linear remnants with intact native vegetation supported more bat species, as did those adjacent to unsealed, as opposed to sealed roads. Fields of unimproved native pastures, with more retained scattered trees and associated hollows and logs, supported the greatest bat species richness and activity. We conclude that the juxtaposition of linear remnants of intact vegetation and scattered trees in fields, coupled with less-intensive land uses such as unimproved pastures will benefit bat communities in agricultural landscapes, and should be incorporated into agri-environment schemes. In contrast, sealed roads may act as a deterrent. The “wildlife friendly farming” vs “land sparing” debate has so far primarily focussed on birds, but here we have found evidence that the integration of both approaches could particularly benefit bats.     

Diversity and community composition of butterflies and odonates in an ENSO-induced fire affected habitat mosaic: a case study from East Kalimantan, Indonesia

Little is known about the diversity of tropical animal communities in recently fire‐affected environments. Here we assessed species richness, evenness, and community similarity of butterflies and odonates in landscapes located in unburned isolates and burned areas in a habitat mosaic that was severely affected by the 1997/98 ENSO (El Niño Southern Oscillation) event in east Kalimantan, Indonesian Borneo. In addition related community similarity to variation in geographic distance between sampling sites and the habitat/vegetation structure Species richness and evenness differed significantly among landscapes but there was no congruence between both taxa. The species richness of butterflies was, for example, highest in sites located in a very large unburned isolate whereas odonate species richness was highest in sites located in a small unburned isolate and once‐burned forest. We also found substantial variation in the habitat/vegetation structure among landscapes but this was mainly due to variation between unburned and burned landscapes and variation among burned landscapes. Both distance and environment (habitat/vegetation) contributed substantially to explaining variation in the community similarity (beta diversity) of both taxa. The contribution of the environment was, however, mainly due to variation between unburned and burned landscapes, which contained very different assemblages of both taxa. Sites located in the burned forest contained assemblages that were intermediate between assemblages from sites in unburned forest and sites from a highly degraded slash‐and‐burn area indicating that the burned forest was probably recolonised by species from these disparate environments. We, furthermore, note that in contrast to species richness (alpha diversity) the patterns of community similarity (beta diversity) were highly congruent between both taxa. These results indicate that community‐wide multivariate measures of beta diversity are more consistent among taxa and more reliable indicators of disturbance, such as ENSO‐induced burning, than univariate measures.     

The diversity of insectivorous bat assemblages among habitats within a subtropical urban landscape

We investigated the bat (Microchiroptera) diversity of four major habitat types within a large Australian subtropical city (Brisbane, Australia) to determine whether species richness was affected by habitat changes associated with urbanization, as suggested from studies elsewhere. Forty sites, ten in each habitat type (remnant bushland, parkland, low‐density residential and high‐density residential) were surveyed using acoustic bat detectors on six non‐consecutive occasions. Fourteen bat species were recorded. The species accumulation curve of the entire Brisbane bat assemblage reached a plateau at 14 species. The total numbers of species in bushland, parkland, low‐density residential and high‐density residential habitats were 14, 13, 14 and 11 species, respectively. Asymptotic estimates of species richness for each habitat were close or equal to these totals. Mean asymptotic estimated species richness differed significantly among habitats, being lowest in high‐density residential sites and highest in low‐density residential sites. Evenness profiles were similar across habitats, and were not strongly dominated by a few species. Partitioning of diversity components showed that landscape (γ) diversity was mainly determined by the high species richness of low‐density residential and bushland habitats (α diversity), rather than high beta (β) diversity among habitats. These findings contradict those of other studies on bat diversity in which species richness was highest within ‘natural’ areas of the urban landscape and assemblages were dominated by one or two species. This highlights the need for caution in making generalizations based on existing information, which is dominated by studies in temperate regions.     

Adaptability and vulnerability of high flying Neotropical aerial insectivorous bats to urbanization

Aim Urbanization is a dominant demographic trend throughout the world that involves massive habitat alterations. Understanding how urbanization affects biota is a crucial prerequisite for development and application of effective species conservation programmes. Our study focuses on Neotropical high flying aerial insectivorous bats, an ecologically important, but so far seriously understudied group of vertebrates. Location Panama. Methods Using acoustic monitoring, we assessed and compared species occurrence, composition and activity of aerial insectivorous bats at three site categories located on the isthmus in Panama: forest, urban areas and a forest–town interface. Results In 2 years of field work, we recorded 44,744 bat passes over the microphone and identified a total of 25 aerial insectivorous bat species. Species richness was highest in the forest, decreased towards the forest–town interface and was lowest at the urban sites, while dominance (Berger‐Parker‐Index) increased from the forest to the urban sites. Overall, general bat activity (passes min^?1) was highest at the forest–town interface and lowest at the urban sites. Multivariate analysis suggests compositional differences in species occurrence and activity among site categories with mainly molossid species occurring in urban areas. Main conclusions Our results clearly demonstrate species‐specific differences between high flying aerial insectivorous bats concerning their adaptability and vulnerability to urban areas. Our results suggest that a suite of morphological traits including species mobility determine persistence of aerial insectivorous species in cities. Our results underline the necessity for detailed assessments of species‐specific habitat requirements and dynamics of species occurrence and activity over time to develop meaningful conservation tools targeted at aerial insectivorous bats.     

Optimizing ultrasonic sampling effort for monitoring forest bats

Landscape‐scale monitoring is a key approach for assessing changes in indicators. However, great care needs to be taken to collect rigorous data and avoid wasting resources in long‐term programmes. Insect‐eating bats are diverse, functionally important and are often proposed as indicator species of environmental health. We used acoustic (ultrasonic) data from pilot bat surveys undertaken in forests and woodlands to optimize sampling effort to produce precise estimates of bat activity and occupancy. We also carried out simulations to evaluate the statistical power of different sampling designs to detect changes in activity and occupancy levels of individual bat species. There was little gain in precision for estimates of bat activity by sampling beyond five to six detector nights. To ensure spatial heterogeneity was sampled around a monitoring point, three detectors for two nights or two detectors for three nights would be required. This level of sampling was also sufficient to be 90% certain of recording occupancy for 11 of 12 taxa. Power simulations revealed that a sampling design using two detectors per monitoring point for two nights could detect a 30% decline within 10 years with 90% power for all species, except the white‐striped free tail bat (Tadarida australis), using either changes in activity levels or occupancy. However, fewer years were required when using occupancy. Setting detectors either on‐flyways or off‐flyways contributed only minor differences to the time taken to reach 90% power for both occupancy and activity levels, though sampling both locations has major implications for interpreting trends in bats. We suggest that bat activity levels are more sensitive for detecting change than occupancy because one pass or 1000 passes can be recorded per night by an acoustic detector, and this is not differentiated by occupancy. Bats can be monitored cost‐effectively and should be included in monitoring programmes.     

Detecting impacts of non‐native species on associated invertebrate assemblages depends on microhabitat

Invasive plants that displace native floral communities can cause changes to associated invertebrate species assemblages. Using a mini‐review of the literature and our own data we add to the still considerable debate about the most effective methods for testing community‐level impacts by invasive species. In endangered saltmarshes of southeast Australia, the non‐native rush Juncus acutus L. is displacing its native congener J. kraussii Hochst., with concurrent changes to floral and faunal assemblages. In two coastal saltmarshes, we tested the hypothesis that the ability to detect differences in the invertebrate assemblage associated with these congeneric rushes depends on the microhabitat of the plant sampled. We used three sampling methods, each targeting specific microhabitats: sweep netting of the plant stems, vacuum sampling of the plant tussock, and vacuum sampling of the ground directly below the plants. Over 3800 individuals and 92 morphospecies were collected across four main taxa: gastropods, crustaceans, hexapods and arachnids. Detection of differences in invertebrate density, richness and composition associated with native compared with non‐native rushes was dependent on the microhabitat sampled and these differences were spatially variable. For example, at one saltmarsh the stems and tussock of J. acutus had a lower density and richness of total invertebrates and hexapods than those of the native J. kraussii. In contrast, crustaceans on the ground were in greater abundance below J. acutus than J. kraussii. This study demonstrates that on occasions where overall differences in the assemblage are not detected between species, differences may become apparent when targeting different microhabitats of the plant. In addition, separately targeting multiple microhabitats likely leads to a greater probability of detecting impacts of invasion. Comparing the invertebrate assemblage without differentiating between or sampling an array of microhabitats can fail to determine the impact of invasive species. These results highlight that a combination of methods targeting different microhabitats is important for detecting differences within the invertebrate community, even for phylogenetically related species.     

Conservation of biodiversity in coffee agroecosystems: a tri-taxa comparison in southern Mexico

We compare species richness of birds, fruit-feeding butterflies and ground-foraging ants along a coffee intensification gradient represented by a reduction in the number of species of shade trees and percentage of shade cover in coffee plantations. We sampled the three taxa in the same plots within the same period of time. Two sites were selected in the Soconusco region of the state of Chiapas, Mexico. Within each site four habitat types were selected and within each habitat type four points were randomly selected. The habitat types were forest, rustic coffee, diverse shade coffee, and intensive coffee (low density of shade). We found different responses of the three taxa along the intensification gradient. While ants and butterflies generally decrease in species richness with the decrease of shade cover, birds declined in one site but increased in the other. Ant species richness appears to be more resistant to habitat modification, while butterfly species richness appears to be more sensitive. Bird species richness was correlated with distance from forest fragments but not with habitat type, suggesting that scale and landscape structure may be important for more mobile taxa. For each of these taxa, the rustic plantation was the one that maintained species richness most similar to the forest. We found no correlation between the three taxa, suggesting that none of these taxa are good candidates as surrogates for each other. We discuss the implications of these results for the conservation of biodiversity in coffee plantations, in particular, the importance of distinguishing between different levels of shade, and the possibility that different taxa might be responding to habitat changes at different spatial scales.     

Biome transitions as centres of diversity: habitat heterogeneity and diversity patterns of West African bat assemblages across spatial scales

It is widely accepted that species diversity is contingent upon the spatial scale used to analyze patterns and processes. Recent studies using coarse sampling grains over large extents have contributed much to our understanding of factors driving global diversity patterns. This advance is largely unmatched on the level of local to landscape scales despite being critical for our understanding of functional relationships across spatial scales. In our study on West African bat assemblages we employed a spatially explicit and nested design covering local to regional scales. Specifically, we analyzed diversity patterns in two contrasting, largely undisturbed landscapes, comprising a rainforest area and a forest‐savanna mosaic in Ivory Coast, West Africa. We employed additive partitioning, rarefaction, and species richness estimation to show that bat diversity increased significantly with habitat heterogeneity on the landscape scale through the effects of beta diversity. Within the extent of our study areas, habitat type rather than geographic distance explained assemblage composition across spatial scales. Null models showed structure of functional groups to be partly filtered on local scales through the effects of vegetation density while on the landscape scale both assemblages represented random draws from regional species pools. We present a mixture model that combines the effects of habitat heterogeneity and complexity on species richness along a biome transect, predicting a unimodal rather than a monotonic relationship with environmental variables related to water. The bat assemblages of our study by far exceed previous figures of species richness in Africa, and refute the notion of low species richness of Afrotropical bat assemblages, which appears to be based largely on sampling biases. Biome transitions should receive increased attention in conservation strategies aiming at the maintenance of ecological and evolutionary processes.     

Efficiency of a Rapid Assessment of the Diversity of Ground Beetles and Ants, in Natural and Disturbed Habitats of the Nahuel Huapi Region (NW Patagonia, Argentina)

We use sample-based rarefaction curves to evaluate the efficiency of a rapid species richness assay of ground beetles and ants captured in pitfall traps in the Nahuel Huapi National Park (NW Patagonia, Argentina). We ask whether ant species richness patterns show some concordance with those of beetles, and use several extrapolation indices for estimating the expected number of species at a regional scale. A total of 342 pitfall traps were spread in groups, at an intensity of 9 traps/100 m², with two collection stations, at each of 19 sites representative of burned and unburned habitats in the forest, scrub and steppe, along a west-to-east transect of 63 km long. The high regional habitat heterogeneity along the west-to-east gradient is paralleled by a turnover of beetle and ant species, although different families of Coleoptera show idiosyncratic responses across habitat types. Spatial stratification of sampling over three major habitats along with the inclusion of burned and unburned environments may improve sampling efficiency. The observed and extrapolated species richness suggests that we captured a high proportion of the total number of species of beetles and ants known for the region. However, trends in species richness of ants may not indicate similar trends in beetles. Ants and beetles cannot be used as surrogate taxa for the analysis of species richness patterns. Instead, both taxa should be considered as focal as they may offer complementary information for the analysis of the effect of disturbance and regional habitat heterogeneity on species diversity patterns at a regional scale.     

Local vs. landscape controls on diversity: a test using surface-dwelling soil macroinvertebrates of differing mobility

Aim A better understanding of the processes driving local species richness and of the scales at which they operate is crucial for conserving biodiversity in cultivated landscapes. Local species richness may be controlled by ecological processes acting at larger spatial scales. Very little is known about the effect of landscape variables on soil biota. The aim of our study was to partly fill this gap by relating the local variation of surface‐dwelling macroarthropod species richness to factors operating at the habitat scale (i.e. land use and habitat characteristics) and the landscape scale (i.e. composition of the surrounding matrix). Location An agricultural landscape with a low‐input farming system in Central Hesse, Germany. Methods We focused on five taxa significantly differing in mobility and ecological requirements: ants, ground beetles, rove beetles, woodlice, and millipedes. Animals were caught with pitfall traps in fields of different land use (arable land, grassland, fallow land) and different habitat conditions (insolation, soil humidity). Composition of the surrounding landscape was analysed within a radius of 250 m around the fields. Results Factors from both scales together explained a large amount of the local variation in species richness, but the explanatory strength of the factors differed significantly among taxa. Land use particularly affected ground beetles and woodlice, whereas ants and rove beetles were more strongly affected by habitat characteristics, namely by insolation and soil characteristics. Local species richness of diplopods depended almost entirely on the surrounding landscape. In general, the composition of the neighbouring landscape had a lower impact on the species richness of most soil macroarthropod taxa than did land use and habitat characteristics. Main conclusions We conclude that agri‐environment schemes for the conservation of biodiversity in cultivated landscapes have to secure management for both habitat quality and heterogeneous landscape mosaics.     

Long‐term effects of repeated fuel‐reduction burning and logging on bats in south‐eastern Australia

Fire regimes have a major influence on biodiversity in many ecosystems around the globe, yet our understanding of the longer‐term response of fauna is typically poor. We sampled bats with ultrasonic detectors in three different years in dry sclerophyll forests of south‐eastern Australia in a long‐term, management‐scale experiment. Frequent low‐intensity burning (every 2 or 4 years plus unburnt) and logging (with 33% retention of the original unlogged tree basal area) were manipulated to assess their effects on bats. We found that both the fire regime and regrowth after logging influenced the local bat community. The routine burning treatment (burnt every 4 years) in unlogged forest was consistently related to higher total bat activity (2–3 times) and species richness when compared to unburnt controls and logging treatments. Foraging activity was more variable, but it was typically lowest in Unlogged Unburnt Controls. These patterns were evident at both the detector site scale and the block scale and were probably due to a reduction in understorey stem density with burning, especially in unlogged forest. Bat activity was significantly lower across the entire study area (including controls) in 1 year, when sampling occurred within 6 months of burning. When pooled across burning treatments, unlogged forest supported higher bat activity (1.5 times) and species richness than logged forest (12‐ to 17‐year‐old regrowth), again most likely because of a negative association with high stem density in regrowth after logging. We conclude that low‐intensity burning had positive benefits for echolocating bats, most notably in unlogged forest. However, careful planning is required to generate heterogeneous vegetation patterns that are likely to be most suitable for a range of taxa.     

ORIGINAL ARTICLE: Are small sedentary species affected by habitat fragmentation? Local vs. landscape factors predicting species richness and composition of land molluscs in Swedish conservation forests

Aim To investigate the relative role of local versus landscape factors for local species diversity of snails and slugs in conservation forests. In landscapes with small, isolated patches of semi‐natural habitats, many species that require large habitat areas have disappeared or are threatened. We asked whether small sedentary taxa that depend on local conditions, such as molluscs, are affected if total habitat area decreases in the landscape. Location Temperate broadleaved and oak‐rich forest in southern Sweden. Methods We sampled molluscs in 25 small conservation forests that are well‐spaced out over a large region. In each forest, sampling was conducted in two plots, each of 1 ha, separated by about 25–100 m. Factors potentially influencing local diversity of molluscs were measured in the plots and in the surrounding landscape at different scales (in space and time) and were analysed by stepwise multiple regression and ordination (PCA and NMS). Results We recorded 53 species, and mean species richness per forest (plots pooled) was 22.6. The pH of the plant litter predicted both species richness and composition; other local (plot) factors of lower importance were canopy openness, stony ground and tree species. The area of conservation forest (woodland key habitat) within 10 km of plots was positively associated with species richness, and was also related to species composition. Openness of the landscape (agriculture) was a negative factor, but historical plot openness (1938–59) seemed to be unimportant. In addition, climate/topography (temperature and altitude) also predicted species composition of the sites. Main conclusions We rejected the hypothesis that microhabitat factors alone, or mainly, determine local species richness and composition of land molluscs. These representatives of small, sedentary organisms seem to be substantially influenced by the surrounding landscape, which should be considered in conservation work and in plans for the protection of forest biodiversity.     

Establishing the evidence base for maintaining biodiversity and ecosystem function in the oil palm landscapes of South East Asia

The conversion of natural forest to oil palm plantation is a major current threat to the conservation of biodiversity in South East Asia. Most animal taxa decrease in both species richness and abundance on conversion of forest to oil palm, and there is usually a severe loss of forest species. The extent of loss varies significantly across both different taxa and different microhabitats within the oil palm habitat. The principal driver of this loss in diversity is probably the biological and physical simplification of the habitat, but there is little direct evidence for this. The conservation of forest species requires the preservation of large reserves of intact forest, but we must not lose sight of the importance of conserving biodiversity and ecosystem processes within the oil palm habitat itself. We urgently need to carry out research that will establish whether maintaining diversity supports economically and ecologically important processes. There is some evidence that both landscape and local complexity can have positive impacts on biodiversity in the oil palm habitat. By intelligent manipulation of habitat complexity, it could be possible to enhance not only the number of species that can live in oil palm plantations but also their contribution to the healthy functioning of this exceptionally important and widespread landscape.     

Ensemble composition and activity levels of insectivorous bats in response to management intensification in coffee agroforestry systems

Shade coffee plantations have received attention for their role in biodiversity conservation. Bats are among the most diverse mammalian taxa in these systems; however, previous studies of bats in coffee plantations have focused on the largely herbivorous leaf-nosed bats (Phyllostomidae). In contrast, we have virtually no information on how ensembles of aerial insectivorous bats--nearly half the Neotropical bat species--change in response to habitat modification. To evaluate the effects of agroecosystem management on insectivorous bats, we studied their diversity and activity in southern Chiapas, Mexico, a landscape dominated by coffee agroforestry. We used acoustic monitoring and live captures to characterize the insectivorous bat ensemble in forest fragments and coffee plantations differing in the structural and taxonomic complexity of shade trees. We captured bats of 12 non-phyllostomid species; acoustic monitoring revealed the presence of at least 12 more species of aerial insectivores. Richness of forest bats was the same across all land-use types; in contrast, species richness of open-space bats increased in low shade, intensively managed coffee plantations. Conversely, only forest bats demonstrated significant differences in ensemble structure (as measured by similarity indices) across land-use types. Both overall activity and feeding activity of forest bats declined significantly with increasing management intensity, while the overall activity, but not feeding activity, of open-space bats increased. We conclude that diverse shade coffee plantations in our study area serve as valuable foraging and commuting habitat for aerial insectivorous bats, and several species also commute through or forage in low shade coffee monocultures.     

Assemblage-level responses of phyllostomid bats to tropical forest fragmentation: land-bridge islands as a model system

Aim Working within a system of high structural contrast between fragments and the surrounding matrix, we assessed patterns of species loss and changes in species composition of phyllostomid bats on artificial land‐bridge islands relative to mainland assemblages, and evaluated the responses of bats to forest edges. We further examined the relative influence of local‐scale characteristics (e.g. vegetation structure, island area) versus landscape attributes (e.g. forest cover, patch density) and the importance of spatial scale in determining phyllostomid species richness and composition on islands. Location Islands in Gatún Lake and adjacent mainland peninsulas in the Barro Colorado Nature Monument, Panama. Methods Bats were sampled over a 2‐year period on 11 islands as well as at forest‐edge and interior sites on adjacent mainland, resulting in > 8400 captures. Results The islands harboured a less diverse and structurally simplified phyllostomid bat fauna. Islands far from the mainland were especially species‐poor. This decline in species richness was associated with compositional shifts towards assemblages strongly dominated by frugivores with good dispersal abilities. Members of other ensembles, most importantly gleaning animalivores, were much less common or absent. Although overall species composition was not significantly altered, species richness at continuous forest‐edge sites was significantly lower compared with that at interior sites. Distance from the mainland and amount of forest cover in the landscape were the best predictors of species richness and assemblage composition. Responses were scale‐dependent. At the local scale, species richness was independent of island area but was correlated positively with distance from the mainland. In contrast, area effects became more important at larger spatial scales, suggesting that many species use multiple fragments. Main conclusions Our results underline the conservation value of small habitat remnants, which, even when embedded in a hostile matrix, can support a relatively diverse bat fauna, provided that there is a low degree of patch isolation and spatial proximity to larger tracts of continuous forest. Although the results at the assemblage level were inconclusive, we demonstrate that certain bat species and ensembles, particularly gleaning animalivores, exhibit high edge‐sensitivity. Our results point to habitat loss rather than changes in landscape configuration as the main process after isolation underlying phyllostomid bat responses, suggesting that conservation efforts should focus on habitat preservation instead of trying to minimize fragmentation per se at the expense of habitat amount.