Prospects for biodiversity conservation in the Atlantic Forest: Lessons from aging human-modified landscapes

Recent global assessments have shown the limited coverage of protected areas across tropical biotas, fuelling a growing interest in the potential conservation services provided by anthropogenic landscapes. Here we examine the geographic distribution of biological diversity in the Atlantic Forest of South America, synthesize the most conspicuous forest biodiversity responses to human disturbances, propose further conservation initiatives for this biota, and offer a range of general insights into the prospects of forest species persistence in human-modified tropical forest landscapes worldwide. At the biome scale, the most extensive pre-Columbian habitats across the Atlantic Forest ranged across elevations below 800 masl, which still concentrate most areas within the major centers of species endemism. Unfortunately, up to 88% of the original forest habitat has been lost, mainly across these low to intermediate elevations, whereas protected areas are clearly skewed towards high elevations above 1200 masl. At the landscape scale, most remaining Atlantic Forest cover is embedded within dynamic agro-mosaics including elements such as small forest fragments, early-to-late secondary forest patches and exotic tree monocultures. In this sort of aging or long-term modified landscapes, habitat fragmentation appears to effectively drive edge-dominated portions of forest fragments towards an early-successional system, greatly limiting the long-term persistence of forest-obligate and forest-dependent species. However, the extent to which forest habitats approach early-successional systems, thereby threatening the bulk of the Atlantic Forest biodiversity, depends on both past and present landscape configuration. Many elements of human-modified landscapes (e.g. patches of early-secondary forests and tree monocultures) may offer excellent conservation opportunities, but they cannot replace the conservation value of protected areas and hitherto unprotected large patches of old-growth forests. Finally, the biodiversity conservation services provided by anthropogenic landscapes across Atlantic Forest and other tropical forest regions can be significantly augmented by coupling biodiversity corridor initiatives with biota-scale attempts to plug existing gaps in the representativeness of protected areas.     

Habitat fragmentation reduces genetic diversity and connectivity among toad populations in the Brazilian Atlantic Coastal Forest

Tropical rainforests are becoming increasingly fragmented and understanding the genetic consequences of fragmentation is crucial for conservation of their flora and fauna. We examined populations of the toad Rhinella ornata, a species endemic to Atlantic Coastal Forest in Brazil, and compared genetic diversity among small and medium forest fragments that were either isolated or connected to large forest areas by corridors. Genetic differentiation, as measured by F_ST, was not related to geographic distance among study sites and the size of the fragments did not significantly alter patterns of genetic connectivity. However, population genetic diversity was positively related to fragment size, thus haplotype diversity was lowest in the smallest fragments, likely due to decreases in population sizes. Spatial analyses of genetic discontinuities among groups of populations showed a higher proportion of barriers to gene flow among small and medium fragments than between populations in continuous forest. Our results underscore that even species with relatively high dispersal capacities may, over time, suffer the negative genetic effects of fragmentation, possibly leading to reduced fitness of population and cases of localized extinction.     

The Brazilian Atlantic Forest: How much is left, and how is the remaining forest distributed? Implications for conservation

The neotropical Atlantic Forest supports one of the highest degrees of species richness and rates of endemism on the planet, but has also undergone a huge forest loss. However, there exists no broad-scale information about the spatial distribution of its remnants that could guide conservation actions, especially when systematic biodiversity data are not available. In this context, our objectives were to quantify how much of the forest still remains, and analyze its spatial distribution. We considered the entire Brazilian Atlantic Forest, and eight sub-regions, defined according to species distribution. The results revealed a serious situation: more than 80% of the fragments are <50 ha, almost half the remaining forest is <100 m from its edges, the average distance between fragments is large (1440 m), and nature reserves protect only 9% of the remaining forest and 1% of the original forest. On the other hand, our estimates of existing Atlantic Forest cover were higher than previous ones (7-8%), ranging from 11.4% to 16%. The differences among estimates are mainly related to our inclusion of intermediate secondary forests and small fragments (<100 ha), which correspond to approximately 32-40% of what remains. We suggest some guidelines for conservation: (i) large mature forest fragments should be a conservation priority; (ii) smaller fragments can be managed in order to maintain functionally linked mosaics; (iii) the matrix surrounding fragments should be managed so as to minimize edge effects and improve connectivity; and (iv) restoration actions should be taken, particularly in certain key areas. The clear differences in the amount remaining and its spatial distribution within each sub-region must be considered when planning for biodiversity conservation.     

Influence of forest fragmentation on an endangered large-bodied lemur in northwestern Madagascar

Madagascar’s diverse and mostly endemic fauna and flora suffer from recent landscape changes that are primarily caused by high levels of human activities. The loss and fragmentation of forest habitats are well known consequences of human activities. In this study, we investigate the effects of forest fragmentation on presence, abundance and genetic diversity in a larger-bodied lemur species, Lepilemur edwardsi, in northwestern Madagascar. In addition, we characterized the genetic differentiation among populations and demographic changes. We found L. edwardsi at only 13 (76.5%) of 17 visited sites, 11 of which were situated in the Ankarafantsika National Park (ANP). We captured between two and 17 individuals per site. We sequenced the mtDNA d-loop of all samples and genotyped 14 microsatellite loci in two exemplary populations for demographic analyses. A negative influence on forest fragmentation could be detected, since the fragments had a lower genetic diversity than sites in the ANP. Genetic differentiation between populations ranged from low to high but was almost always significant. A typical pattern of isolation-by-distance could not be detected and the data could rather be interpreted as results of random genetic drift. The data furthermore revealed signals of a demographic collapse of about two orders of magnitude in the two exemplary sites. This decline probably started during the last few hundred years of intensified human disturbances and population growth. Given the results of this study, urgent conservation actions are needed and should concentrate on an effective protection of the few remaining populations in order to ensure the long-term survival of L. edwardsi.     

Extreme population divergence and conservation implications for the rare endangered Atlantic Forest sloth, Bradypus torquatus (Pilosa: Bradypodidae)

The maned sloth (Bradypus torquatus) is an endangered species endemic to the Atlantic Forest of eastern Brazil. This biome has been reduced to 7% of its original extent and the remaining forests are highly fragmented. We analyzed 70 samples from the largest remnant populations in the states of Bahia, Espírito Santo and Rio de Janeiro to characterize their geographic structure and to produce estimates of genetic diversity. The analysis indicated that the remnant populations are reproductively isolated and extremely divergent. The populations present a very discontinuous distribution, with divergent genetic clusters specific to different geographical regions, probably caused by allopatric fragmentation. This pattern is likely related to Pleistocenic climatic and vegetation changes, and indicates the presence of at least two independent evolutionary units. The analyses also indicate that populations separated by more than 100 km should be considered different management units. Thus, devastation of the Atlantic Forest leads to an unrecoverable loss of genetic diversity in this species. These conclusions should direct conservation actions aimed at preserving the distinctiveness of each evolutionary unit, as well as to preserve the demographic isolation of different management units.     

Plant communities at the periphery of the Atlantic rain forest: Rare-species bias and its risks for conservation

Initiatives that establish species rarity as an indicator of conservation priority might be biased if they disregard important evolutionary and adaptive processes taking place in lower diversity communities and ecotones. Conservation policies regarding the Atlantic forest strongly emphasize the core formation (i.e. the rainforest stricto sensu) rather than the marginal habitats (e.g. restingas, swamps, and high altitude campos) and species that are rare/endemic. To discuss this issue I revisit a hypothesis I have forwarded in 2002 that postulates that plant colonization of habitats marginal to the Atlantic rain forests of the State of Rio de Janeiro was largely related to terrestrial nurse plants that originally, in the rainforest habitat, were canopy plants such as epiphytes or hemi-epiphytes. Adaptations to water and nutrient restrictions, typical of life in the canopy, granted success to such plants upon migration to sandy, swampy or rocky substrates in neighbouring areas. Many such species, then, behaved as nurse plants and favoured colonization of these more extreme habitats by a number of other rainforest species. I now review recent evidence that corroborate this hypothesis, while examining the nature of such nurse plants. In all marginal habitats, nurse plants are often highly abundant locally and have high ecophysiological vigour, while both widespread and endemic species are found among them. Thus, nursing effect, local abundance, and ecophysiological performance are not related to species geographic distribution or to their spectrum of habitat preference. Paradoxically, several abundant nurse plant species have low Darwinian fitness. These studies provoke two reflections. First, the Atlantic forest sensu lato, i.e. the core formation plus the peripheral ones, should be treated collectively as a biodiversity hotspot, rather than the core rainforest formation alone. Second, widespread or common species play important functional roles in such marginal habitats and, despite their ubiquity, ecologically they might be less fit than rare/endemic ones at the local level due, for instance, to current constraints to sexual reproduction. Thus, they should also be targeted as conservation priorities.     

On the restoration of high diversity forests: 30 years of experience in the Brazilian Atlantic Forest

We present a review of more than 30 years of ecological restoration in the Brazilian part of the Atlantic Forest. Based on what has been done in this biome, we try to summarize the main findings and challenges for restoration in this highly threatened forest biome. We found that many past experiences did not result in self-perpetuating forests, for different reasons. Currently, most projects aim to construct self-sustaining communities and no longer see restoration as a deterministic process. We also found that the reconstruction of permanent forest with high diversity is feasible but it depends on the strategies applied and on the surrounding landscape. Although many new techniques have been created (e.g. seed rain management or promotion of natural regeneration), the most used one in the Atlantic Forest is still the planting of many native species from different functional groups. Native species are largely used and perform well even in highly disturbed environments. Today, many projects are trying to produce thousands of hectares of permanent forests and many technical advances are about to be incorporated. But restoration also faces some main challenges to become an effective and widespread means of conserving the Atlantic Forest which are, namely, reducing costs, planning restoration actions at landscape-level, and conforming to socio-political issues. The socio-political tools to overcome such barriers in practice have yet to be developed.     

Arbuscular mycorrhizal fungi in the Brazilian Atlantic forest: A gradient of environmental restoration

Arbuscular mycorrhizal fungi (AMF) are of great importance for the successful regeneration of degraded natural areas. The objective of this study was to examine how the time of environmental recuperation is affecting the occurrence and diversity of AMF species in riparian areas belonging to the Atlantic Forest biome in the State of São Paulo, Brazil. The study involved a native forest area (NT) and a gradient of environmental restoration: five (R05), ten (R10), and twenty (R20) years after reforestation. Soil samples were collected in the rainy (January) and dry season (June). Chemical, physical and microbiological analyses were performed including the amount of glomalin and quantification of AMF spores. The frequency of occurrence of genera and ecological indices, as richness (R), Shannon's diversity (H) and Simpson's dominance index (Is) were calculated. The largest spore number was found in R05 and the highest richness and diversity indices of AMF species in NT. Considering the two sampling periods and the four areas studied, we found 22 AMF species, and the genera Glomus and Acaulospora were the most frequent. A Canonical Discriminant Analysis showed that Glomus viscosum, Acaulospora scrobiculata, Acaulospora mellea and Scutellospora heterogama were the species that contributed the most to distinguishing the areas. Moisture, density and glomalin were positively correlated with the number of spores, however, soil nitrate showed a negative correlation. This work gives a better understanding of the interactions between AMF and forest soils and allows to know the distribution of AMF species according to environmental recovery time.     

Bacterial diversity and community structure along different peat soils in boreal forest

Bacteria in peat forest soil play important role in global carbon cycling. The distribution of bacteria population in different peat soils as a whole and how forest management practices alter the bacterial populations are still poorly known. Using pyrosequencing analysis of 16S rRNA gene, we quantified the diversity and community structure of bacteria in eight peat forest soils (pristine and drained) and two mineral forest soils from Lakkasuo, Finland with either spruce-dominant or pine-dominant tree species. In total, 191,229 sequences which ranged from 15,710 to 22,730 per sample were obtained and affiliated to 13 phyla, 30 classes and 155 genera. The peat forest soils showed high bacterial diversity and species richness. The tree species seems to have more strong impact on the bacterial diversity than the type of peat soil, which drives the changes in bacterial community structure. The dominant taxonomic groups across all soils (>1% of all sequences) were Proteobacteria, Acidobacteria, Actinobacteria, Bacteroidetes, Planctomycetes and Verrucomicrobia. The relative abundance of bacteria phylum and genus differed between soil types and between vegetation. Significant differences in relative abundance of bacteria phyla were only found for Gemmatimonadetes and Cyanobacteria between the pristine and the drained peat forest soils. At genus level, the relative abundance of several genera differed significantly between the peat soils with same or different tree species, including Burkholderia, Caulobacter, Opitutus, Mucilanginibacter, Acidocella, Mycobacterium, Bradyrhizobium, Dyella and Rhodanobacter.     

Carabid beetle (Coleoptera: Carabidae) diversity in the black spruce succession of eastern Canada

This study provides the first assessment of carabid beetle diversity in a natural forest context that encompasses a complete black spruce (Picea mariana) natural succession. Boreal forest conservation has been based on several assumptions about forest age that only consider species richness without accounting for species composition. It has also been guided by studies of incomplete chronologies that do not include naturally burned or old-growth stages. Twenty-one forest stands of different ages following fire - from recently burned to old-growth stages (0-340 years of age) - were sampled, revealing a strong relationship between age of forest and diversity of Carabidae. Over time, species richness followed a parabolic U-shaped pattern both with observed (Obs) and predicted richness (ACE). Chronological clustering identified four groups of species in the succession: the ‘burned’ group characterized 0-2-year-old forests, ‘regenerating’ (21-58 year), ‘mature’ (70-170 year) and ‘old-growth’ (177-340 year). The time spans corresponding to each of these assemblages lengthen with age of forest at an exponential rate. Ward’s and K-means (clustering without constraint) provided support for the four assemblages but showed variation between individual successions, particularly for the ‘regenerating’ assemblage, identified as the most heterogeneous. The IndVal method identified characteristic species in every stage of the succession, particularly Sericoda spp. in the burned stage and Dromius piceus and Platynus mannerheimii associated with old-growth stands. The results obtained here show that diversity of Carabidae varies in primeval conditions according to age of forest and such variation should be taken into account when conservation issues are involved.     

Long-term erosion of tree reproductive trait diversity in edge-dominated Atlantic forest fragments

Habitat loss and fragmentation promote relatively predicable shifts in the functional signature of tropical forest tree assemblages, but the full extent of cascading effects to biodiversity persistence remains poorly understood. Here we test the hypotheses that habitat fragmentation (a) alters the relative contribution of tree species exhibiting different reproductive traits; (b) reduces the diversity of pollination systems; and (c) facilitates the functional convergence of reproductive traits between edge-affected and early-secondary forest habitats (5-32 years old). This study was carried out in a severely fragmented 670-km² forest landscape of the Atlantic forest of northeastern Brazil. We assigned 35 categories of reproductive traits to 3552 trees (DBH ? 10 cm) belonging to 179 species, which described their pollination system, floral biology, and sexual system. Trait abundance was calculated for 55 plots of 0.1 ha across four habitats: forest edges, small forest fragments (3.4-83.6 ha), second-growth patches, and core tracts of forest interior within the largest available primary forest fragment (3500 ha) in the region. Edge-affected and secondary habitats showed a species-poor assemblage of trees exhibiting particular pollination systems, a reduced diversity of pollination systems, a higher abundance of reproductive traits associated with pollination by generalist diurnal vectors, and an elevated abundance of hermaphroditic trees. As expected, the reproductive signature of tree assemblages in forest edges and small fragments (edge-affected habitats), which was very similar to that of early second-growth patches, was greatly affected by both habitat type and plot distance to the nearest forest edge. In hyper-fragmented Atlantic forest landscapes, we predict that narrow forest corridors and small fragments will become increasingly dominated by edge-affected habitats that can no longer retain the full complement of tree life-history diversity and its attendant mutualists.     

From genetic diversity and structure to conservation: Genetic signature of recent population declines in three mouse lemur species (Microcebus spp.)

The exceptional biodiversity of Madagascar is threatened by anthropogenic landscape changes that took place during the 2000 years of human colonization. This study focuses on the influence of geographic distance and forest fragmentation on genetic diversity and population differentiation of three rare, nocturnal, arboreal lemur species in northwestern Madagascar. Historic declines in population sizes as a consequence of forest fragmentation are quantified and dated. Eighteen sites were visited, and a total of 205 Microcebus ravelobensis, 45 M. bongolavensis and 78 M. danfossi were genotyped with eight microsatellite loci. Genetic differentiation among the sites, as measured by F_ST, ranged from 0.01 to 0.19. These values were significant in almost all cases and indicated genetic structure in the samples. Isolation-by-distance was detected in one species and a STRUCTURE analysis indicated that fragmentation further promoted genetic differentiation. Bayesian methods revealed that populations from all three species underwent a major demographic collapse of around two orders of magnitude. This decrease probably began after the arrival of humans, most likely within the last 500 years. This result suggests that anthropogenic changes may have been limited during the first 1500 years of human colonization in all three ranges. Two of the study species (M. danfossi, M. bongolavensis) lack effectively protected areas in their ranges. Consequently, quick conservation actions are now needed in order to secure the remaining genetic diversity of these species.     

Forest loss with urbanization predicts bird extirpations in Vancouver

Forest loss with accompanying urbanization is often permanent as forests are replaced by impervious concrete structures or surfaces. The remaining forest patches are usually isolated and too small to maintain viable populations of forest-dependent wildlife species. Consistent with the species-area function (S = cA^z), extirpation of species should follow forest loss. However, this has rarely ever been tested in the urban landscape. We determined whether the observed number of forest-dependent bird extirpations occurring after forest loss matches that predicted by the species-area function in Vancouver (study area ∼126.7 km²), British Columbia, Canada. Prior to European settlement in 1859, Vancouver was covered completely by coastal western hemlock forest. Based on published historical maps of forest cover and present Landsat imagery of Vancouver, we estimated that approximately 87% of the forest area in Vancouver has been converted to urban development. Using the species-area function, we then predicted that approximately 14 bird species closely associated with lowland forest should have been extirpated. This figure is significantly greater than the three bird species known to have been extirpated in Vancouver. When we constrained the comparison to bird species closely associated with lowland forest and restricted in their distribution in British Columbia, we found a close match between the observed and predicted number of extirpations, suggesting that bird species closely associated with lowland forest and restricted in their geographic distribution are most prone to extirpation as a result of forest loss. Given that urban areas are situated in productive habitats with inherently high species richness, it is important that local land-use plans incorporate the conservation of habitat fragments from the onset and place particular emphasis on the regionally restricted forest-dependent species.     

Functional diversity of microbial communities in the mixed boreal plain forest of central Canada

This study was designed to examine whether or not specific tree species (Picea glauca, Picea mariana, Pinus banksiana, Populus tremuloides), their post-fire stand age, or their position in a successional pathway had any significant effect on the functional diversity of associated soil microbial communities in a typical mixed boreal forest ecosystem (Duck Mountain Provincial Forest, Manitoba, Canada). Multivariate analyses designed to identify significant biotic and/or abiotic variables associated with patterns of organic substrate utilization (assessed using the BIOLOG™ System) revealed the overall similarity in substrate utilization by the soil microbial communities. The five clusters identified differed mainly by their substrate-utilization value rather than by specific substrate utilization. Variability in community functional diversity was not strongly associated to tree species or post-fire stand age; however, redundancy analysis indicated a stronger association between substrate utilization and successional pathway and soil pH. For example, microbial communities associated with the relatively high pH soils of the P. tremuloides-P. glauca successional pathway, exhibited a greater degree of substrate utilization than those associated with the P. banksiana-P. mariana successional pathway and more acidic soils. Differences in functional diversity specific to tree species were not observed and this may have reflected the mixed nature of the forest stands and of their heterogeneous forest floor. In a densely treed, mixed boreal forest ecosystem, great overlap in tree and understory species occur making it difficult to assign a definitive microbial community to any particular tree species. The presence of P. tremuloides in all stand types and post fire stand ages has probably contributed to the large amount of overlap in utilization profiles among soil samples.     

Biodiversity conservation in human-modified Amazonian forest landscapes

Amazonia (sensu lato) is by far the largest tropical forest region, but has succumbed to the highest absolute rates of tropical deforestation and forest degradation, driven by rapid frontier expansion, road-building, and spontaneous or government-subsidized migration. The large area-through-time and paleo-climatic stability of Amazonian forests may help explain the high regional to local scale plant and animal species diversity of true forest specialists and high ecological sensitivity to contemporary land-use change. We describe the prevailing forms of anthropogenic disturbance that affect forest organisms in the context of the geographic and evolutionary background that has shaped the degree to which forest species may be resilient to environmental change. The fate of Amazonian biodiversity will partly depend upon the interaction between land-use and climate change, and the extent to which seasonally-dry forests can retain immunity against catastrophic recurrent wildfires. This review illustrates the importance of considering interactions between different forms of forest disturbance to develop effective conservation policy. We conclude with some considerations of the policy agenda necessary to protect forest cover and forest biodiversity at a meaningful scale across the Amazonian biome.     

Effect of forest fragmentation on fruit and seed predation of the tropical dry forest tree Ceiba aesculifolia

Forests fragmentation reduces the density of natural plant populations forming patches of the remaining individuals. One of the biotic interactions that can be affected by forest fragmentation and is poorly studied is seed predation. We determined the effects of forest fragmentation on seed and fruit predation in Ceiba aesculifolia by comparing trees in continuous forest with trees in fragmented forest. We compared the following variables: (a) frequency of fruit predation by Collie’s squirrel (Sciurus colliaei) in each habitat; (b) frequency of the cotton-staining bug seed predator (Dysdercus, Orden Hemiptera) in each habitat; (c) the effect of seed predation on germination frequency and time; and (d) the effect of different life stages of Dysdercus on seed viability. In continuous habitat, 100% of the trees presented fruits with squirrel predation while only 34% of trees in fragmented habitats presented fruit predation. In continuous forest 27% of the trees contained fruits with the seed predator Dysdercus, while only 2% of the trees in fragmented forest presented Dysdercus. The initial weight of damaged seeds was greater than seeds that were not damaged indicating that seed predators select heavier seeds to feed upon. Frequency of seed germination was affected by different life stages; pre-adults decreased germination significantly more than nymphs and adults. Seed predation significantly increased the time it took for germination to occur. Our study shows that forest fragmentation significantly affects predation patterns of squirrels and cotton-staining bugs. Reduction of natural seed predators in forest fragments may have long-term consequences on forest structure and diversity.     

Effect of litter quality on its decomposition in broadleaf and coniferous forest

Recently there has been much interest in the effect of litter mixing as well as the effect of different forest habitats on the decomposition process. Our aim was to test two hypotheses: high quality litter promotes decomposition of poor quality litter, and litter decomposes faster in broadleaf than in coniferous forest. We conducted a litter mixing experiment using litterbags placed in two forest floors, in which treatments consisted of litter monocultures of each of two campy species (Castanopsis eyrei and Pinus massoniana), as well as mixtures of these two species. The results showed that C. eyrei leaves decomposed significantly faster in the coniferous habitat than in their native habitat. On the other hand, P. massoniana needles decomposed significantly faster in their native coniferous habitat than in the broadleaf habitat. In our experiment we found that the mixture had different effect on different quality litter. P. massoniana needles (poor quality) had a positive effect on the decomposition of C. eyrei leaves (high quality), while C. eyrei leaves had a negative effect on the decomposition of P. massoniana needles in the mixture case in both broadleaf and coniferous habitats. The diversity of the fungi identified from different litters varied among treatments and the mass loss was positively correlated with the Shannon–Weaver diversity index of fungi. It is suggested that fungi may be one of the major drivers to control the decomposition process.     

Avian conservation priorities in a top-ranked biodiversity hotspot

Rio de Janeiro state in Brazil has one of the most diverse and most endangered avifaunas in the continental Americas. Many of these endangered birds are endemic to the Atlantic Forest biodiversity hotspot, and some even endemic to Rio de Janeiro itself. As with all other forested hotspots, little original forest remains. Much of that is outside formal protected areas and faces the risk of deforestation. These factors create special circumstances for setting conservation priorities — ones common to hotspots in general — but typically not to many conservation priority setting exercises.We mapped the distribution of the remaining habitat for the 189 birds in Rio de Janeiro state that are officially endangered and/or endemic to the Atlantic Forest. Using those habitat maps, we calculated the amount of habitat currently within protected areas for each species. We then prioritized all non-protected parts of the state for their avian conservation value and their potential contribution to a comprehensive protected area system. This analysis identified 10% of the remaining unprotected part of the state as the highest priority for avian conservation. We further highlight specific locations where conservation actions could create a more comprehensive protected area system for the avifauna of Rio de Janeiro state.     

Differential effect of arbuscular mycorrhizal fungal communities from ecosystems along management gradient on the growth of forest understorey plant species

Arbuscular mycorrhizal fungi (AMF) are important functional components of ecosystems. Although there is accumulating knowledge about AMF diversity in different ecosystems, the effect of forest management on diversity and functional characteristics of AMF communities has not been addressed. Here, we used soil inoculum representing three different AM fungal communities (from a young forest stand, an old forest stand and an arable field) in a greenhouse experiment to investigate their effect on the growth of three plant species with contrasting local distributions - Geum rivale, Trifolium pratense and Hypericum maculatum. AM fungal communities in plant roots were analysed using the terminal restriction fragment length polymorphism (T-RFLP) method. The effect of natural AMF communities from the old and young forest on the growth of studied plant species was similar. However, the AMF community from the contrasting arable ecosystems increased H. maculatum root and shoot biomass compared with forest inocula and T. pratense root biomass compared to sterile control. According to ordination analysis AMF inocula from old and young forest resulted in similar root AMF communities whilst plants grown with AM fungi from arable field hosted a different AMF community from those grown with old forest inocula. AMF richness in plant roots was not related to the origin of AMF inoculum. G. rivale hosted a significantly different AM fungal community to that of T. pratense and H. maculatum. We conclude that although the composition of AM fungal communities in intensively managed stands differed from that of old stands, the ecosystem can still offer the ‘symbiotic service’ necessary for the restoration of a characteristic old growth understorey plant community.