Altered leaf-litter decomposition rates in tropical forest fragments

The effects of forest fragmentation on leaf-litter decomposition rates were investigated for the first time in an experimentally fragmented tropical forest landscape in Central Amazonia. Leaf-litter decomposition rates were measured at seven distances (0–420 m) along forest edge-to-interior transects in two 100-ha fragments, two continuous forest edges, and at an identical series of distances along two deep continuous forest transects, as well as at the centers of two 1-ha and two 10-ha fragments. Decomposition rates increased significantly towards the edge of 100-ha forest fragments. Litter turnover times were 3–4 times faster within 50 m of the edge of 100-ha fragments than normally found in deep continuous forest. In contrast, there was no significant change in the rate of leaf-litter decomposition from the interior to the edge of continuous forest. It is difficult to account for these very different edge responses. Decomposition rates were not correlated with air temperature differentials, evaporative drying rates, litter depth, biomass or moisture content, or with total invertebrate densities, either within individual edge transects or across all sites. The difference in edge response may be due to chance, particularly the patchy removal of vast quantities of litter by litter-feeding termites, or may be a real, area-dependent phenomenon. Clearly, however, forest fragmentation increases the variability and unpredictability of litter decomposition rates near forest edges. In addition to edge effects, decomposition rates were strongly affected by decreasing fragment area. While sites at the centers of 10-ha and 100-ha forest fragments and continuous forest had equivalent decomposition rates, rates were markedly lower at the centers of 1-ha fragments. Litter turnover times were 2–3 times slower in 1-ha fragments than in continuous forest, and up to 13 times slower than at 100-ha edges. Litter structure and nutrient cycling dynamics are inevitably altered by forest fragmentation. Received: 16 October 1997 / Accepted: 14 April 1998     

Vegetation and microclimatic edge effects in two mixed-mesophytic forest fragments

Forest edges are known to consist of microenvironments that may provide habitat for a different suite of species than forest interiors. Several abiotic attributes of the microenvironment may contribute to this change across the edge to center gradient (e.g., light, air temperature, soil moisture, humidity). Biotic components, such as seed dispersal, may also give rise to changes in species composition from forest edge to interior. We predicted that abiotic and biotic measures would correlate with distance from forest edge and would differ among aspects. To test these predictions, we measured abiotic and biotic variables on twelve 175 m transects in each of two 24 ha forest fragments in east-central Illinois that have remained in continuous isolation for upwards of 100 years. Both univariate and multivariate techniques were used to best describe the complex relationships among abiotic factors and between abiotic and biotic factors. Results indicate that microclimatic variables differ in the degree to and distance over which they show an edge effect. Relative humidity shows the widest edge, while light and soil moisture have the steepest gradients. Aspect influences are evidenced by the existence of more pronounced edge effects on south and west edges, except when these edges are protected by adjacent habitat. Edges bordered by agricultural fields have more extreme changes in microclimate than those bordered by trees. According to PCA results, species richness correlates well with microclimatic variation, especially light and soil moisture; however, in many cases species richness had a different depth of edge influence than either of these variables. The herbaceous plant community is heavily dominated by three species. Distributions of individual species as well as changes in plant community composition, estimated with a similarity index, indicate that competition may be influencing the response of the vegetation to the edge to interior gradient. This study indicates that edge effects must be considered when the size and potential buffering habitat of forest preserves are planned.     

Age distribution of trees in stationary forest system

A statistical theory for the age distribution of spatially dominant trees in a stationary forest system is developed. The result depends whether or not mortality is spatially correlated, as well as whether or not the stand boundaries are pre-determined. In the case of spatially non-correlated mortality, the tree age distribution is an exponential with survival rate as the base. In the case of spatially correlated mortality within a stand with pre-determined boundaries, the age distribution within the stand is an exponential with natural base. For a small stand, the median life span of the stand is inversely proportional to the number of trees (n); the median life span in relation to stand closure time is inversely proportional to n ln(n). For a large stand, the stand life does not extend to the closure time.The behaviour of a forest system without fixed stand boundaries depends on the dimensionality of the system. In the case of a one-dimensional system, the longevity distribution is exponential, most of the trees however having the same longevity. Consequently, the probability density of tree age is constant. However, the probability mass of size of catastrophe destroying a particular tree is evenly distributed. This is due to trees being rapidly born on empty areas in the beginning of the life cycle, and clusters rapidly growing into larger ones close to the end of tree life.     

Ground nest predation might not be higher along edges of Neotropical forest remnants surrounded by pastures: evidence from the Brazilian Atlantic forest

Several studies indicate that nest predation is higher along edges than in habitat interiors mainly due generalist predators arising from or proliferating in the surrounding matrix. Recent reviews demonstrate however that this is far from universal, in part because studies are strongly biased in temperate regions. Far fewer are known from the Neotropics and just a handful of studies have been carried out in the biologically-rich but severely fragmented Atlantic Forest of Brazil. Here we tested the influence of edge proximity on ground nest predation in a large (21,787 ha) Atlantic forest reserve. The experiment was carried out using chicken and quail eggs in 12 transects with 500 m in length, half of which parallel to internal edges (dirt roads) and half parallel to external edges (forest/pasture). Nest predation was significantly higher in wet season (42.7%), when no difference was found between edge and forest interior, than in dry season (16.5%), when nest predation was higher in forest interior (400 m). Within seasons, the difference between internal and external edges and the association between edge distance and edge type were not significant. Results suggest that ground nest predation in large protected areas of the Atlantic forest is mainly caused by forest dwellers rather than by species inhabiting the surrounding grassland-dominated landscape, mirroring recent findings in other tropical areas.     

Tree species diversity in small, tropical riparian forest fragments in Belize, Central America

Tree species diversity was measured in a network of very small galleryforests within the Mountain Pine Ridge savanna in Belize. Research focussed onforest patches smaller than 1 ha in size (micro-forests) and linearstrips of trees along creeks lacking interior core zones with low understoreylight levels (tree thickets). Twenty-five micro-forests and 51 tree thicketsites were sampled throughout the savanna. A total of 144 morphospecies 5cm dbh (106 in micro-forests and 117 in tree thickets) werefound, which represents 1/5 of the approximately 700 native tree species in Belize.Most (85.3%) of the species encountered are typically found in tropical rainforests and few are restricted to savanna or riparian environments. Speciesaccumulated at a much faster rate in micro-forests than in tree thickets. Onlyone species, the palm Acoelorraphe wrightii, was extremelyabundant, accounting for almost 30% of all stems. Many of the species werepresent in very low densities: 19% of all species found in micro-forests and 42%of those found in tree thickets had on average one or fewer stems per hectare. Alarge proportion of species were also found infrequently across the landscape,being present in only 36% of micro-forests and at 52% of tree thicket sites. Theresults indicate that networks of very small forest patches can contain highnumbers of species and could therefore contribute to the maintenance of regionalbiodiversity.     

Spatio-temporal patterns of tree community dynamics in a tropical forest fragment in South-east Brazil

The tree community (dbh > 5 cm) of a fragment of tropical montane semi-deciduous forest in South-east Brazil was repeatedly surveyed over a 19-year period in order to assess spatial and temporal patterns of dynamics. The surveys took place in 1987, 1992, 1996, 2001, and 2006 in a grid of 126 20 × 20 m permanent plots covering almost the entire fragment (5.8 ha). Overall patterns indicated that a self-thinning process has taken place in the fragment since 1992. Community dynamics varied in space and time, with most dynamics highly spatially clustered. With exception of mortality rates, there were no changes in the spatial patterns of community dynamics through time. No relation between edges and dynamics variables was found. Most species with increasing density and basal area were shade-bearers, while most decreasing species were canopy light demanders and pioneers.     

Edge and shape effects on ant (Hymenoptera: Formicidae) species richness and composition in forest fragments

In this paper, we tested four hypotheses relative to edge and shape effects on ant communities: (i) forest edges have lower species richness than the remnant core; (ii) species richness increases with distance from the edge; (iii) irregularly shaped remnants have lower species richness than more regular remnants; (iv) there is a higher similarity of species composition between edge and core in irregular than in regular remnants. We sampled litter ant communities on the edge and core of ten remnants, in Viçosa, Minas Gerais, Brazil. Species richness was larger at the forest core than at the edges, although did not increase with distance from the edge. Species richness did not vary with shape complexity. The similarity of species composition between edge and core showed a decreasing trend with remnant area, and did not vary with shape complexity. The observed differences of species richness between forest core and edge may be due to higher harshness of edges, caused by environmental changes. The absence of relationship between species richness and distance from the edge might indicate the range of edge effects, which would be smaller than the smallest distance of core sampled. Therefore, edges would affect litter-dwelling ant species richness in a distance smaller than 50 m. The observation of species composition allowed us to notice an effect of fragmentation that would not be noticed if we were considering only species richness. Edge may serve as step to generalist species, which may use it to colonise forest remnants. Furthermore, small remnants are more colonisation-prone by such species, and have a more homogeneous species composition than large remnants.     

Population structure and spatial patterns for trees in a temperate old-growth evergreen broad-leaved forest in Japan

The population structure and spatial pattern of major tree species in a warm-temperate old-growth evergreen broad-leaved forest in the Tatera Forest Reserve of Japan were investigated. All stems 5 cm in diameter at breast height (DBH) were mapped on a 4 ha plot and analyses were made of population structure and the spatial distribution and spatial association of stems in different vertical layers for nine species. This was done in the context of scale dependency. The plot was located on a very gentle slope and 17.1% of its canopy layer was in gaps. It contained 45 woody plant species and 4570 living stems with a basal area of 63.9 m² ha^–1. Castanopsis cuspidata var. sieboldii, the most dominant species for the basal area, had the maximum DBH among the species present, fewer smaller stems and a lower coefficient of statistical skewness of the DBH distribution. The second most dominant species, Dystilium racemosum, had the highest stem density (410 ha^–1), more abundant smaller stems and a relatively higher coefficient of skewness. Most stems in different vertical layers showed a weakly aggregated distribution with loose colonies as basic units. Gap dependency for the occurrence of stems under the canopy layer was weak. Maximum slope degree of the plot also weakly affected the occurrence of stems. Spatial associations varied among intra- and interspecific cohorts in the different layers and spatial scales examined, and positive associations among cohorts were found more frequently as the scales examined became larger. This tendency suggests that key factors forming observed spatial associations might vary with the spatial scales.     

Growth strategies of main trees and forest architecture of aFagus-Magnolia forest in Florida, USA

Growth strategies of six species of trees are compared and used to analyze forest architecture. They included the overstory speciesFagus grandifolia, Magnolia grandiflora, Pinus glabra andLiquidambar styraciflua, and the understory speciesOstrya virginiana andIlex opaca. The six species were abundant in Woodyard Hammock, an old-growth forest in northern Florida, USA. Height, stem diameter, crown projection and radial growth were measured in trees between 5 and 35 m tall. Three different, but non-exclusive, growth strategies were identified for the tree species: survival (Fagus grandifolia, Magnolia grandiflora, Ilex opaca), occupy open space (Fagus grandifolia, Ostrya virginiana, Ilex opaca), and reach above competitors (Liquidambar styraciflua, Pinus glabra). In two transects (0.42 ha) and one quadrat (1 ha), heights of dominant trees were used to distinguish different phases of forest development, which were mapped. In the quadrat, juvenile canopy trees in the undergrowth were mapped. The combination of presence of different developmental phases, presence of juveniles in the undergrowth, growth strategies of main tree species, and disturbance regime was used to assess forest development in the near future.Fagus grandifolia is predicted to become the main dominant species, but the frequent hurricanes open the forest canopy and provide opportunities for understory species (Ostrya virginiana andIlex opaca), and light-demanding overstory species (Liquidambar styraciflua andPinus glabra).     

Effects of forest fragmentation on small mammals in an Atlantic Forest landscape

Recent studies on the effects of tropical forest fragmentation indicate that fragmented landscapes are complex and heterogeneous systems influenced by factors other than the size or degree of isolation of forest remnants: of particular importance are the quality of the matrix and the edge-induced habitat changes. In order to investigate the influence of these factors, small mammals were surveyed in 36 sites in the landscape of Una, a region that encompasses some of the last and largest Atlantic Forest remnants in northeastern Brazil. Six sites were distributed on each of six landscape components – the interiors and edges of small remnants, the interiors and edges of large remnants, and the most common forested habitats found in the matrix. The survey comprised 46,656 trap-nights and yielded 1725 individuals of 20 species of rodents and marsupials. Results revealed: an increase in beta-diversity caused by fragmentation; the contrasting effects of the altered forested habitats of the matrix, which harbor both forest and disturbance-adapted species; a greater importance of edge effect than of patch size to the observed changes in small mammal community in remnants; an association among terrestrial forest species and among arboreal forest species in terms of the distribution and abundance in the Una mosaic; and a distinctive vulnerability of these two groups of species to fragmentation. Results emphasize the biological importance and conservation value of both fragmented landscapes and small remnants in the Atlantic Forest, as well as the importance of management techniques to control and attenuate edge effects.     

Effect of fragmentation on the community structure of epixylic bryophytes in Atlantic Forest remnants in the Northeast of Brazil

The effects of fragmentation and edge effects on the floristic composition, richness, diversity and abundance of epixylic bryophytes (growing on decaying wood) were investigated in ten fragments of Atlantic Forest remnants in the Northeast of Brazil. In each fragment, four perpendicular 100 m transects were demarcated. Along these transects, samples of bryophytes growing on decaying wood were collected. The forest fragments were grouped in three size classes (small: <100 ha; medium: 100–500 ha; large: >500 ha). Correlation and multivariate analysis were undertaken between bryophyte flora and fragment metrics (size, form, isolation, altitude variation, nuclear area and secondary vegetation percentage and distance from the edge). A total of 99 species of bryophytes, 52 liverworts and 47 mosses were registered. The statistical results confirming fragment size is an important factor in epixylic community structure. Therefore, composition, richness, diversity and abundance can be better explained by a junction of all studied landscape factors. Bryophyte richness, the percentage of samples with the greatest coverage of decaying wood and shade-tolerant species’ distribution, were not correlated to distance from the forest edge. This suggests that edge effects are not linear or can be detected beyond 100 m from the edge, which is very important for inclusion in future studies.     

Identifying spatial patterns in the tropical rain forest structure using hemispherical photographs

This work emphasises the usefulness of hemispherical photography for identifying spatial patterns in the tropical rain forest structure. Structural variability was investigated at a local (intra-site) scale, in relation to the forest mosaic, and at a regional (inter-sites) scale, for its implication in forest typology. Four primary forest sites, from 0.6 to 1 ha, were investigated in Central Sumatra, Indonesia.In a first instance, a qualitative analysis of the forest 3-D structure was found to be very helpful in the interpretation of quantitative results related to forest dynamics. The quantitative analysis was undertaken through the assessment of three structural characteristics: (1) the canopy openness (CO), or visible sky as seen from all directions of the hemisphere, (2) the spherical variance (SV), which quantifies the spatial dispersion of gaps and, (3) the plant area index (L_p), defined as half the surface area of canopy elements per unit ground area. At the local scale, maps of the CO values gave an indication about disturbance location and extent, providing an interesting document for studies on forest dynamics. At the regional scale, between-sites comparisons of CO, SV and L_p values added new information on forest structural differences when compared to dendrometric measurements. These results emphasised the fact that rain forest may exhibit high structural variability, even within a same bioclimatic region and a narrow altitude range. Hemispherical photographs could be a quick means of further investigating this spatial variability and it's relation to physical environment, thus providing information that is crucial for the refining of forest typology in the area.     

Decomposition and carbon cycling of dead trees in tropical forests of the central Amazon

Decomposition rate constants were measured for boles of 155 large dead trees (>10 cm diameter) in central Amazon forests. Mortality data from 21 ha of permanent inventory plots, monitored for 10–15 years, were used to select dead trees for sampling. Measured rate constants varied by over 1.5 orders of magnitude (0.015–0.67 year^–1), averaging 0.19 year^–1 with predicted error of 0.026 year. Wood density and bole diameter were significantly and inversely correlated with rate constants. A tree of average biomass was predicted to decompose at 0.17 year^–1. Based on mortality data, an average of 7.0 trees ha^–1 year^–1 died producing 3.6 Mg ha^–1 year^–1 of coarse litter (>10 cm diameter). Mean coarse litter standing-stocks were predicted to be 21 Mg ha^–1, with a mean residence time of 5.9 years, and a maximum mean carbon flux to the atmosphere of 1.8 Mg C ha^–1 year^–1. Total litter is estimated to be partitioned into 16% fine wood, 30% coarse wood, and 54% non-woody litter (e.g., leaves, fruits, flowers). Decomposition rate constants for coarse litter were compiled from 20 globally distributed studies. Rates were highly correlated with mean annual temperature, giving a respiration quotient (Q ₁₀) of 2.4 (10°C^–1). Received: 14 June 1999 / Accepted: 31 August 1999     

Dung beetles in continuous forest, forest fragments and in an agricultural mosaic habitat island at Los Tuxtlas, Mexico

With the aim of determining what kind of landscape mosaics might sustainmaximum diversity and minimum species loss, dung beetles were sampled withbaited pitfall traps to compare species richness and species composition in atract of continuous forest, forest fragments and a habitat island consisting ofa mosaic of forest and arboreal crops in Los Tuxtlas, southern Mexico. Wecaptured 7332 dung beetles representing 33 species. Similar numbers of specieswere captured in the three habitats. However, 56% of individuals were capturedin the continuous forest, 29% in the mosaic habitat and 15% in the forestfragments. Eight species (Canthon femoralis,Copris laeviceps, Canthidium centrale,Onthophagus batesi, Deltochilumpseudoparile, O. rhinolophus, Canthonviridis vazquezae and Dichotomius satanus) accounted for 90% of thecaptures, but their relative dominance varied among habitats. A clear trend wasevident in the number of dung beetles captured in the dung processing guilds(rollers/tunnelers) as well as in the diurnal and nocturnal guilds, withcaptures decreasing from the continuous forest to the mosaic habitat to theforest fragments. A similar trend was detected in detection rates for medium andsmall size dung producing mammals. Species richness of forest fragments andmosaic habitat did not differ from that found in the continuous forest, butthese habitats differ significantly in species richness from isolated shaded andunshaded plantations, linear strips of vegetation, the forest–pasture edge andpastures according to rarefaction analysis. The co-occurrence of the continuousforest, the mosaic habitat and the cluster of forest fragments in closeproximity seems to be preserving a diverse assemblage of dung beetle species inthe local landscape.     

Strong influence of long-distance edge effect on herb-layer vegetation in forest fragments in an agricultural landscape

In agricultural landscapes in central Europe, species richness of the herbaceous plant community may be compromised by processes associated with forest fragmentation, habitat loss, and management practices. We examined variability in species richness and composition of the herbaceous layer in 229 plots located in 23 forest fragments (0.1 to 255 ha), in a representative upland agricultural landscape in central Bohemia, in relation to the most important site environmental factors, edge effects, and site history. The influence of environmental factors on the composition of vegetation in the herb layer was evaluated using generalized additive models, which enabled us to analyze highly non-linear and non-monotonic relationships. Total species richness and number of red-listed and ancient forest species were significantly influenced by type of forest vegetation, light quality, soil pH, slope aspect, and distance from the forest edge. Implications of the significant explanatory variables corresponded well to previous findings, with the exception of distance from the forest edge, for which we found a positive relationship with species richness for distances up to 200 m toward the forest interior. Plant species with low colonization ability occupied plots with increasing frequency from edge to forest interior, while fast-colonizing species showed the opposite trend. Apart from the edge effect, forest continuity should be considered for its important contribution to the richness of ancient forest and red-listed species, whereas the effect of forest fragment size appeared to be generally weak. These results do not negate the importance of large forest fragments for the maintenance of herb layer species richness, but specifically emphasize the essential contribution of the core habitats of these forests. In summary, we showed that the negative effects of habitat fragmentation on the richness of ancient forest and red-listed species and on herb layer species in total can be largely attributed to either the edge effect itself or to aggregate effects of forest edge and forest continuity.     

Spatial distribution patterns of trees at different life stages in a warm temperate forest

We have investigated tree distributions in relation to topography between different tree life history stages, from the seed-dispersal stage to the adult stage in a warm temperate evergreen broadleaved forest on Yakushima Island, Japan, to clarify the critical stages in determining adult tree distributions. We conducted a census of all living trees > or =30 cm tall and collected seed falls over three years using 25 seed traps in a 50 m x 50 m quadrat. Four life stages were defined: stage 1, dispersed seed; stage 2, individuals taller than 30 cm and diameter at breast height (DBH) < 1 cm; stage 3, trunks 1 cm < or = DBH < 10 cm; stage 4, trunks with DBH > or = 10 cm. We classified 17 common tree species into three groups; group A was distributed mainly on the upper slope, group B on the lower slope, and group C on both. Most of group A and B trees at stages 2-4 showed an aggregated distribution along the topographical gradient. The densities at stage 1 showed weaker aggregations according to slope. Topography-specific tree distribution was probably determined at the regeneration stage, and later survival was less effective as a mechanism of vegetation differentiation.     

Effects of forest-pasture edge on C,N and P associated with <Emphasis Type="Italic">Caesalpinia eriostachys</Emphasis>, a dominant tree species in a tropical deciduous forest in Mexico

Forest fragmentation in tropical ecosystems can alter nutrient cycling in diverse ways. We have analysed the effects of the forest-pasture edge on nutrient soil dynamics in a tropical deciduous forest (TDF) in Mexico. In two remnant forest fragments, both larger than 10 ha, litterfall, litter and soil samples associated to the tree Caesalpinia eriostachys were collected at five distances from the pasture edge into the inner forest (10 m in the pasture and 0–10, 30–40, 70–80 and 100–110 m towards the forest interior). We measured the concentrations of carbon (C), nitrogen (N) and phosphorus (P) in litterfall, surface litter and soil, and soil microbial C (C_mic) and microbial N (N_mic). Soil nutrient concentrations and C_mic and N_mic were lower in the pasture soils than in the forest soil samples. Total C and N pools, and C_mic and N_mic in the pasture were lower than in the forest. In contrast, net N immobilization and the increase in N_mic from rain to dry season increased from the edge to the inner forest. Soil P concentration was lower in the pasture and at the first distance class in the forest margin (0–10 m) than in the sites located further into the forest, while litter P concentration had the inverse pattern. Litterfall P was also reduced near the edge and increased towards the forest interior. As a consequence, litterfall C:P and N:P ratios decreased from the edge to the inner forest. These results suggest that the forest–pasture edge disrupts P dynamics within the first 10 m in the forest. Thus, plants' use of nutrients and productivity could be altered in the edge of fragmented forests.     

Diversity of Neotropical migratory landbird species assemblages in forest fragments and man-made vegetation in Los Tuxtlas,Mexico

We investigated the presence of Neotropical migratory landbirds in a 90-km² landscape in the region of Los Tuxtlas, Veracruz, Mexico. Using the fixed-radius count point procedure, migratory landbirds were surveyed in 21 forest fragments and in four replicates of shaded (coffee, cacao and mixed) and unshaded (citrus and allspice) plantations, live fences, non-arboreal crops (corn and jalapeño chili pepper) and pastures. The surveys resulted in the count of 4732 birds representing 72 species. While forest fragments accounted for 65% of the total species count, 73% of the birds were counted in the arboreal man-made habitats. Pastures contributed to 10% of the species and to 1% of the individuals counted. Live fences were particularly rich in individuals, accounting for 28% of the birds counted. Rarefaction analysis showed that forest fragments were the sites richest in species, followed by shaded and unshaded plantations and by live fences. Pastures were the habitats poorest in species, followed by non-arboreal crops. Species richness of Neotropical migratory landbirds was associated to vertical and horizontal diversity of vegetation in the habitats investigated. Shaded and unshaded plantations as well as live fences were more similar to forest fragments in species assemblages than non-arboreal crops and pastures. We discuss the conservation value of arboreal agricultural habitat and of live fences in conjunction with forest fragments as temporary habitats for Neotropical migratory landbirds that stop over or winter in Los Tuxtlas.     

A comparison on the response to forest fragmentation by medium-sized Iberian carnivores in central Spain

We studied the use of forest fragments by five medium-sized carnivorespecies in 280 forest fragments on the two Iberian plateaus. We looked forindirect evidence (faeces, tracks, dens) of fragment use (occurrence) by thespecies and analysed whether occurrence could be related to four groups ofvariables: local (vegetation structure and patch size), landscape (distance topossible colonisation sources), regional (fragment location on the northern orsouthern plateau) and the vegetation type of the fragments. We analysed thedifferential response of species according to their life-history and behaviouraltraits. The relationship between use by each species and the factors studied wasanalysed using stepwise logistic regressions. Results indicate that threefactors are crucial to explain fragment use: fragment size, geographic locationand vegetation type. Large fragments are used more than smaller ones, thenorthern plateau is more suitable than the southern plateau, and holm oak andbroad-leaved oak forests are more used than pine woods. The effects andmagnitude varied slightly among species, depending on habitat requirements orlife-history traits. Data indicate that conservation strategies in fragmentedenvironments must take into account elements functioning at different spatialscales, and that it is essential to consider each case within a characteristicregional context.     

Reproductive ecology of tropical forest trees in logged and fragmented habitats in Thailand and Costa Rica

Invertebrates mediate several important ecological processes, including pollination and seed predation, and events that affect invertebrate diversity or behaviour can potentially disrupt forest regeneration processes. This study investigates the impact of logging in Thailand and forest fragmentation in Costa Rica on the pollination and seed production of two self-incompatible forest trees. Logging in a dry deciduous dipterocarp forest in Thailand resulted in reduced densities of the common dipterocarp treeShorea siamensis and variably isolated individual trees. The number of flower visits to S. siamensis by pollinating Trigona bees was not affected by logging disturbance. However, pollinators did spend longer periods of time foraging in the canopies of isolated trees which were more prevalent in logged areas where tree density had been reduced. Consequently, at the logged site few cross-pollinations were effected and fruit set of S. siamensis was considerably lower than at nearby unlogged sites where distances between flowering conspecifics were smaller. Reduced fruit set has long-term implications for the recovery of S. siamensis populations in disturbed areas, and local population genetic structure is likely to be affected as reduced outcrossing rates among trees in disturbed regions results in relatively inbred seed. In Costa Rica forest fragmentation has restricted the once widespread tree Anacardium excelsum to forest patches located in an agriculturally-dominated landscape. As with S. siamensis, the abundance of pollinators, also Trigona bees, in the canopies of A. excelsum was largely unaffected by fragment size. Nevertheless, pollination success and seed production was positively correlated with fragment size. We propose that small bees rarely move between forest fragments and gene exchange through pollination occurs predominantly among trees within fragments and, together with likely low genetic variability in small fragments, that this contributes to the observed reduced fertilisation and seed set of A. excelsum. Thus increased tree isolation tree through selective logging or habitat fragmentation by forest clearance can result in reduced seed set due to changes in the foraging patterns of poorly mobile pollinators. Even if population sizes of the pollinators are maintained following environmental perturbation, this study shows that disturbance may disrupt pollination processes through changes in pollinator foraging behaviour. More attention needs to be focussed on changes in the behaviour of species involved in key ecological interactions following disturbance events in tropical forests.