Two globally threatened tree species, the Critically Endangered Cola porphyrantha (Malvaceae) and the Endangered Gigasiphon macrosiphon (Fabaceae) are narrowly distributed in Kenya and Tanzania. In Tanzania, both species were first located in an isolated, unprotected forest fragment in the East Usambara Mountains in the early 2000s. As no assessment of these subpopulations had been made since then, we surveyed the forest fragment as well as nearby unprotected forest fragments. In contrast to the early 2000s when only five and two mature trees of C. porphyrantha and G. macrosiphon, respectively, were located, we found 18 and five mature trees of these species. We did not find either species in intensive surveys of seven neighbouring unprotected forest fragments but we located a single G. macrosiphon beside a river close to one of the unprotected fragments. Gigasiphon macrosiphon was also previously known from two sites in Amani Nature Reserve in the East Usambara Mountains, but recent surveys, including our own, failed to relocate these subpopulations. Because of heavy anthropogenic disturbance in the one site where the two species still occur and their general absence from adjacent forest, we are working with the local community to protect the isolated fragment. Additionally, in situ planting of locally grown seedlings of both species is being supported by Amani Nature Reserve.

Annona crassiflora is a fruit-bearing tree species native to the Cerrado that has ecological and economic potential, mainly due to the production of attractive and useful fruits for a number of species, including humans. To provide a basis for its conservation and breeding, the objective was to assess the diversity and genetic structure of natural populations of the species using inter-simple sequence repeat (ISSR) molecular markers. Eight populations were analysed in the state of Minas Gerais, Brazil, with 24 individuals randomly sampled in each population, reaching a total of 192 trees. Ten ISSR primers were used, resulting in high genetic diversity for the combined data (H* = 0.35 and I* = 0.52). However, a wide range of representative values was observed in the individual assessment of populations, with JAN, GM and MC standing out for their low genetic diversity, resulting in H* of 0.19, 0.23 and 0.24 and I* of 0.28, 0.34 and 0.35, respectively. An analysis of molecular variance showed greater variation within populations, indicating gene flow (Nm), but genetic differentiation between populations was moderate. Bayesian analysis, although resulting in four genetic groups, revealed the presence of a majority group for the GM and JAN populations. We propose measures to maintain these populations, such as raising awareness of local extractivism and planting genetically divergent seedlings. Furthermore, we recommend including of all populations in conservation and breeding programmes, aiming to cover the maximum genetic variation for the species.

Destruction of tropical rainforests reduces many unprotected habitats to small fragments of remnant forests within agricultural matrices. To date, these remnant forest fragments have been largely disregarded as wildlife habitat, and little is known about mammalian use of these areas in Sumatra. Here, we conducted camera trap surveys (2285 trap-nights) within Bukit Barisan Selatan National Park and five surrounding remnant forest fragments during 2010–2013 and used species composition metrics to compare use. We found 28 mammal species in the protected forest and 21 in the fragments. The fragments harboured a subset of species found in the protected forest and several species not observed in the protected forest. Critically endangered species such as Sunda pangolin (Manis javanica) and Sumatran tiger (Panthera tigris sumatrae) were found in the forest fragments, along with species of conservation concern such as marbled cat (Pardofelis marmorata) and Asiatic golden cat (Pardofelis temminckii). The biodiversity found within the fragments suggests that these small patches of remnant forest may have conservation value to certain mammal species and indicates the importance of further research into the role these habitats may play in landscape-level, multispecies conservation planning.

Seed dispersal effectiveness (SDE) is related to the role of seed dispersal agents in realizing the reproductive potential of plants through seed dispersal and subsequent plant recruitment. The SDE of a given seed disperser may vary spatially and temporally, with important implications for our understanding of the mutualistic relationships involving plants and frugivores. Here we observed 22 frugivorous bird species visiting an individual tree (Cabralea canjerana) over a 9-y period in a fragment of Atlantic forest in south-eastern Brazil to document the temporal variation in SDE. The quantitative (that takes into account the frequency of feeding visits to the focal plant and the number of fruits removed per visit) and qualitative (that considers the probability of seed deposition on a suitable site for plant recruitment) components of SDE varied in two and one order of magnitude both among bird species and among years within bird species, respectively. As a result, the SDE of bird species fluctuates over the years, especially for a migratory species that is the main seed disperser of C. canjerana (Chivi vireo Vireo chivi) whose quantitative component (i.e. the product of visit and fruit intake rates) varied from 3.0 to 7.1. This study illustrates the dynamic nature of the seed-dispersal environment to which a plant is subjected to over the course of its life, which has important consequences for the plant's individual fitness.

Habitat fragmentation creates habitat edges, and ecological edge effects can cause major changes in the ecology and distribution of many taxa. However, these ecological changes may in turn influence animal movements and lead to molecular edge effects and edge-related genetic structure, matters that are largely unexplored. This study aims to infer molecular edge effects and to test three possible underlying mechanisms in the Endangered golden-brown mouse lemur Microcebus ravelobensis, a nocturnal species in the dry deciduous forest of the Ankarafantsika National Park in north-western Madagascar. Mouse lemurs were sampled in one edge and two interior habitats in close proximity to each other (500–1,400 m) in a continuous forest. A total of 41 mouse lemur samples were genotyped with seven nuclear microsatellites, and a fragment of the mitochondrial control region was sequenced for all samples. The overall genetic diversity (allelic richness, heterozygosity, haplotype richness, nucleotide diversity) was lower in the edge habitat compared to the two interior sites and all subpopulations showed signals of relatively low genetic exchange and significant genetic differentiation between them despite the short geographical distances, supporting the local preference model. These findings can be interpreted as preliminary signals of a molecular edge effect and suggest the potential for local adaptation. They are highly relevant for the conservation of fragmented populations, because a further subdivision of already small populations may increase their vulnerability to stochastic demographic changes and collapse.

Forest fragmentation, habitat loss and isolation may have a strong effect on biodiversity in tropical forests. This can include modification of the genetic diversity and structure of plant populations. In this study, we assessed the genetic diversity and structure of the tree Manilkara zapota in 15 naturally formed fragments of semi-evergreen tropical forest, as well as in an adjacent continuous forest for comparison. Forest fragments were scattered within a matrix of wetlands and were highly variable in terms of size and degree of isolation. The naturally fragmented populations of M. zapota had slightly less allelic diversity (Ar: 3.4) than those of the continuous forest (Ar: 3.6), when corrected for sample size. However, populations in the fragments and continuous forest had very similar heterozygosity levels (HE: 0.59 in both cases). Low levels of genetic differentiation were observed among populations (FST: 0.026) and genetic structure was not consistent with isolation by distance, indicating high levels of gene flow. Genetic diversity was not explained by fragment size or degree of isolation. The relatively high genetic diversity and low inter-population genetic differentiation observed in M. zapota may be the result of long-distance pollen and seed dispersal, as well as the high proximity among patches.

Pollen and seed dispersal are key ecological processes, directly impacting the spatial distribution, abundance and genetic structure of plant populations; yet, pollen- and seed-dispersal distances are poorly known. We used molecular markers to identify the parental origin (n = 152 adult trees) of 177 Spondias radlkoferi (Anacardiaceae) seeds deposited by the spider monkey (Ateles geoffroyi) in latrines located beneath 17 sleeping-trees in two continuous forest sites (CF) and two forest fragments (FF) in the Lacandona rain forest, Mexico. We estimated mean parent-offspring (PO) distances per latrine and, for those seeds (54% of seeds) with more than one candidate parent (i.e. the potential maternal and parental parents), we also estimated parent-parent (PP) distances per latrine, and tested if PO and PP distances differed between forest types. Global PO and PP distances per latrine averaged 682 m (range = 83–1741 m) and 610 m (range = 74–2339 m), respectively, and did not differ significantly between CF and FF. This suggests that pollen dispersal is extensive in both forest types and that long seed dispersal distances (>100 m) are common, thus supporting the hypothesis that the spider monkey is an effective seed disperser of S. radlkoferi in continuous and fragmented forests.

Leaf traits of tropical tree species are known to operate as intrinsic determinants of insect herbivory. However, we know little about how habitat fragmentation affects these traits and what, if any, are the consequences of this process on herbivory. We tested the effects of forest fragmentation on the leaf traits of sapling of four light-demanding species: Acalypha diversifolia, Hampea nutricia, Myriocarpa longipes, Siparuna thecaphora, and two shade-tolerant species: Pseudolmedia glabrata and Garcinia intermedia, in Los Tuxtlas, Mexico. We also conducted an acceptability assay with a generalist herbivore Spodoptera frugiperda. Plant traits did not change with forest fragmentation, but did with plant regeneration mode and species identity. Light-demanding species had significantly higher water content, nitrogen concentration and specific leaf area than shade-tolerant species. The latter had significantly higher leaf strength, carbon concentration and carbon:nitrogen ratio. Acceptability was affected by fragmentation but only in P. glabrata; plant tissue from forest fragments was consumed 2.6 times more than that from continuous forest. We conclude that forest fragmentation did not affect leaf traits in this site.

Repeated use of sleeping trees (STs) by frugivores promotes the deposition and aggregation of copious amounts of seed, thus having key implications for seed dispersal and forest regeneration. Seed-rain patterns produced by this behaviour likely depend on the frequency of use of these sites, yet this hypothesis has been poorly tested. We evaluated community-level seed-rain patterns produced by the spider monkey (Ateles geoffroyi) over 13 mo in latrines located beneath 60 STs in the Lacandona rain forest, Mexico. Because this primate is increasingly ‘forced’ to inhabit fragmented landscapes, we tested whether sleeping-tree fidelity (STF) differed among sites and between continuous and fragmented forests. We also tested whether seed-rain patterns were associated with STF within each site and forest type. STF was highly variable among STs (average = 7 mo, range = 1–12 mo), but did not differ among study sites or forest types. STF was positively associated with seed abundance, species diversity and species turnover. Nevertheless, STF tended to be negatively related to seed community evenness. These results are likely due to the most frequently used STs being in areas with greater food density. Our results demonstrate that site fidelity shapes community-level seed-rain patterns and thus has key ecological implications.

Several studies have evaluated the short-term effects of tropical forest fragmentation on plant taxonomic diversity, while only a few have evaluated its effects on functional or phylogenetic diversity. To our knowledge no study has looked at the long-term consequences of tropical forest fragmentation on the three main components of plant diversity simultaneously: taxonomic, functional and phylogenetic diversity. We sampled the vascular flora using belt transects (50 × 4 m) in a continuous tropical semi-evergreen forest (16 transects) and in an adjacent naturally fragmented forest (fragments of 1.7-My-old semi-evergreen forest immersed in a mangrove/sedge matrix) (18 transects), and compared their taxonomic, functional and phylogenetic plant diversity. There were 36 species in the continuous forest and 28 in the fragmented forest. Continuous forest was taxonomically more diverse (25%) than the fragmented forest. All functional diversity metrics were greater (6–33%) in the continuous than in the fragmented forest. Phylogenetic diversity was 19% greater and phylogenetically more overdispersed in the continuous forest than in the fragmented forest. The results suggest that in the fragmented forest not only is taxonomic plant diversity lower, but functional and phylogenetic diversity are as well. The negative effects of forest fragmentation on plant diversity seem to be chronic.

This study was conducted in the Knuckles Forest Region in central Sri Lanka, and investigated how termite species richness, abundance and functional group diversity vary in different montane forest types and identified the likely causes of this pattern. Termite diversity declined with increased elevation, with upper montane forests recording a single endemic species, Postelectrotermes militaris Desneux. Transect sampling in lower montane forests yielded 26 species, with a higher number from dry forests (22 species) than from wet forests (15 species). Species specificity also was high in dry forests (11 species) compared with wet forests (four species). Termite abundance did not show a distinct trend in dry and wet forests. Live-wood termites were present only in upper montane and high-altitude lower montane dry forests. Wet forests had a higher relative abundance (78%) but not species richness (40%) of soil and soil–wood interface feeders. In dry forests, both species richness (82%) and abundance (88%) of fungus-growing wood feeders were higher. The study suggests that key drivers of the species distribution pattern are low temperature and differing forest floor conditions. In the upper montane forest floor where earthworms dominate, wet soil and damp, woody litter riddled with beetles are not favourable for termites. In lower montane wet forests, moist, thick decomposing leaf litter and in dry forests, drier, relatively undecomposed leaf litter with many dry sticks and branches support species with specific food habits.

Overexploitation of forests by humans can lead to highly fragmented populations of forest-dependent species that have poor dispersal abilities. We tested the influence of habitat quality, landscape structure and human pressure on densities of the Barbary macaque Macaca sylvanus in the largest wild population, in the Middle Atlas of Morocco. We surveyed 14 forest fragments of 5–142 km2 that are separated from each other by an inhospitable matrix. We estimated the habitat quality of these forest patches by analysing the vegetation structure and the intensity of human pressure. We studied the distribution of forest patches and estimated macaque densities by combining line-transect sampling and complete group counts. We estimated mean density of individuals to be 9 km−2 (range 0.2–23 km−2). Differentiation of forest patches by a principal component analysis did not show any significant relationship between vegetation type and macaque density. A linear regression model showed that human pressure had a negative impact on density and that density responded positively to patch size. Patch shape, connectivity and altitude did not explain variation in population density. The size of this population is estimated to be c. 5,000 individuals. The survival of several small subpopulations is seriously threatened. As the Middle Atlas is the stronghold of this Endangered species, we recommend modifying forestry practices, reducing overgrazing by livestock within forests, and halting clear-cutting of holm oaks.

Understanding the response of biodiversity to land-use changes is an important challenge for ecologists. We assessed the effects of five landscape metrics (forest cover, number of patches, edge density, mean inter-patch isolation distance and matrix quality) and three patch metrics (patch size, shape and isolation) on the number of species and patch occupancy of medium- and large-sized terrestrial mammals in the fragmented Lacandona rain forest, Mexico. We sampled mammal assemblages in 24 forest patches and four control areas within a continuous forest. The landscape metrics were measured within a 100-ha buffer, and within a 500-ha buffer from the centre of each sampling site. A total of 21 species from 13 families was recorded. The number of species increased with shape complexity and patch size at the patch scale, and with matrix quality within 100-ha landscapes. When considering 500-ha landscapes, only the number of patches (i.e. forest fragmentation level) tended to have a negative influence at the community level. Different landscape and patch metrics predicted the occurrence of each species within the sites. Our results indicate that there is a gradient of tolerance to forest cover change, from highly sensitive species to those tolerant of, or even benefited by, forest-cover change.

We evaluated effects of corridors between forest fragments surrounded by pastures in tropical Mexico. We used experimental translocations and capture–recapture data to measure the proportion of birds returning and time to return after translocation between connected and unconnected patches (five replicates for each treatment). Depending on each species’ degree of forest dependence (forest-restricted and forest-unrestricted species), we assigned birds to two groups to evaluate influence of species characteristics on effects of corridors on movement. Birds translocated between connected patches (n = 75) were seven times more likely to be recaptured in their original capture site when compared with birds translocated between unconnected patches (n = 109). Effects differed among the two species groups. In the presence of corridors, 46% of forest-unrestricted birds returned to the capture site while only 5% returned between unconnected patches. Forest-restricted birds showed similar results, but were only twice as likely to return to a connected capture site. Birds translocated between unconnected patches took longer to return than birds translocated between connected patches. The strong positive effect of corridors on movement, even for forest-unrestricted species, suggests that forested corridors facilitate bird movement and help maintain connectivity even in this highly fragmented landscape.

Secondary forests occupy a growing portion of the tropical landscape mosaic due to regeneration on abandoned pastures and other disturbed sites (Asner et al. 2009). Tropical secondary forests and degraded old-growth forests now account for more than half of the world's tropical forests (Chazdon 2003), and provide critical ecosystem services (Brown & Lugo 1990, Guariguata & Ostertag 2001).

Understanding the effects of forest fragmentation on biodiversity is essential for successful and efficient forest conservation. Four factors may cause loss of biodiversity in forest fragments: the effect of non-random sampling of the original forest, reduced forest size, isolation and edge effects. A review of 58 papers on effects of forest fragmentation reveals that general conclusions from fragmentation research are biased due to a focus on birds, on size-effects rather than isolation, and on species presence rather than population sizes. Perhaps the most important finding is that current knowledge on fragmentation effects is based mainly on studies in small fragments (<10 ha). These are dominated by edge effects, can not contain viable populations for many species and are rarely the focus of conservation programmes. Studies of small fragments can not be extrapolated to larger-sized, protected areas, and do not necessarily support the case for needing extremely large, protected areas. Conservation of medium-sized, strategically-located areas may be a more efficient option for biodiversity conservation, given financial, social and logistic limitations. More research is needed on forest fragments that arc representative of the sizes of real-world protected areas (i.e. >10 000–100 000 ha) and should focus on the biological and human-induced processes which determine species persistence.

The seed dispersal system of a neotropical tree, Cabralea canjerana (Meliaceae), was studied in two forested areas in southeastern Brazil. The first study site, Parque Estadual Intervales (PEI), is a 49,000-ha reserve composed mostly of old-growth Atlantic rain forest. The second site, Mata de Santa Genebra (MSG), is a 250-ha fragment of old-secondary semideciduous forest whose present bird fauna differs markedly from the original, in part as a consequence of forest fragmentation. At PEI 35 bird species ate the diaspores of C. canjerana. Black-tailed tityra (Tityra cayana, Tyrannidae) was the main seed disperser, but several other species were also important seed dispersers. In contrast, at MSG C. canjerana diaspores were eaten by 14 bird species. At this area, the red-eyed vireo (Vireo olivaceus, Vireonidae) was the most important seed disperser, but it was also a ‘waster’ which dropped seeds beneath parent plants, or carried them to sites unsuitable for germination. At PEI, exposed seeds on the forest floor were heavily preyed upon by rodents and insects. Insects destroyed mainly seeds deposited near to parent plants. Insect predation was less intense at MSG than at PEI. The rodent density at MSG was unusually small, and part of the post-dispersal seed predation may be done by terrestrial birds, such as doves and tinamous, which are especially common at MSG. Some of the differences recorded between the seed dispersal systems of C. canjerana at PEI and MSG may have been the result of the fragmentation and isolation of the latter area.

A survey of the diurnal raptor community (46 resident species) at five altitudinal levels from 100 to 4600 m was made in Venezuela using 106 1-km2 sample quadrats and then compared with similar censuses (211 plots) in southwestern Colombia and eastern Ecuador. Six habitat types were defined and habitat preferences were measured from every encountered birds and from the association between individual abundances and the coverage of habitat types on plots. The overall species richness steeply declined with elevation in all three countries. The raptor community consisted mostly of tropical species that rapidly decreased in both abundance and frequency of occurrence with elevation. A small set of subtropical-temperate forest specialists partly replaced them at mid-elevation and only two species occurred in the paramo of Venezuela, against 5–7 in the other countries. Although originally the gradients were probably almost completely covered with forest, a large part of the raptor community is now composed of grassland species, three of them extending from the lowlands to the tree line. The relative impoverishment of the Venezuelan community, especially at upper levels, is attributed to a peninsular effect at the narrow extreme northern tip of the Andes and perhaps also to former forest fragmentation. Within some species, different subspecies had discrete altitudinal distributions and interspecific competition may explain differences in the range extension of some species between countries. Today, however, accelerating deforestation may result in a severe decline of forest-interior species and a probable reduction in the altitudinal range of at least some of them. Species naturally associated with forest gaps or edges usually tolerate a high degree of forest fragmentation and degradation but only one of them may become abundant in heavily human-altered woodlands. Few grassland specialists have become more widely distributed in the increasingly deforested landscapes.

Transformation of tropical forests is likely to affect seed-dispersal patterns and influence the composition of epiphytic plant communities in human-altered habitats. We tested this hypothesis by carrying out a comparative study of seed influx, survival and growth of transplanted seedlings of epiphytic bromeliads among isolated trees in six pasture areas, six forest edges and six forest sites in a montane area in Costa Rica. In total, 72 traps trapped 1285 seeds over a 2-mo period in the dry season of 2003. For all four investigated bromeliad genera, Catopsis, Guzmania, Tillandsia and Werauhia, the number of trapped seeds in each habitat followed a pattern similar to the number of fruiting individuals in the vicinity of the traps. Traps in forest edges (30) were 1.9 times more likely to collect seeds than traps at forest interiors (30) and pasture trees (12), the latter showing similar probabilities of catching seeds. After 1 y, survival and growth of 3660 transplanted seedlings from three bromeliad species was significantly higher in forest interiors, providing no explanation for the lower abundance of fruiting adults in that habitat. These results suggest that the successful establishment of epiphytic bromeliads in forest interiors is mainly dispersal-limited. If corroborated, differences in abundance among species at each habitat are likely related to differences in growth rates and reproductive success. Further studies on the growth and mortality of seedlings up to the flowering stage, however, are needed.