Summary1 Disturbance may cause community composition across sites to become more or less homogenous, depending on the importance of different processes involved in community assembly. In north-eastern North America and Europe local (alpha) diversity of forest plants is lower in forests growing on former agricultural fields (recent forests) than in older (ancient) forests, but little is known about the influence of land-use history on the degree of compositional differentiation among sites (beta diversity). 2 Here we analyse data from 1446 sites in ancient and recent forests across 11 different landscapes in north-eastern North America and Europe to demonstrate decreases in beta diversity and in the strength of species-environment relationships in recent vs. ancient forests. 3 The magnitude of environmental variability among sites did not differ between the two forest types. This suggests the difference in beta diversity between ancient and recent forests was not due to different degrees of environmental heterogeneity, but rather to dispersal filters that constrain the pool of species initially colonizing recent forests. 4 The observed effects of community homogenization and weakened relationships between species distributions and environmental gradients appear to persist for decades or longer. The legacy of human land-use history in spatial patterns of biodiversity may endure, both within individual sites and across sites, for decades if not centuries.

As landscapes throughout Europe and eastern North America recover from past agricultural use, forests continue to reflect their agricultural history. For centuries after agriculture has ceased, plant communities on abandoned agricultural lands remain impoverished in herbaceous species characteristic of uncleared forests. To facilitate the recovery of biological diversity in these forests, and to anticipate the effects of future land-use decisions, we need to understand the process of recolonization. The unique interactions between forest herbs and agricultural history also allow us to explore some universal questions in ecology, such as how dispersal and environment limit species distributions.

As landscapes throughout Europe and eastern North America recover from past agricultural use, forests continue to reflect their agricultural history. For centuries after agriculture has ceased, plant communities on abandoned agricultural lands remain impoverished in herbaceous species characteristic of uncleared forests. To facilitate the recovery of biological diversity in these forests, and to anticipate the effects of future land‐use decisions, we need to understand the process of recolonization. The unique interactions between forest herbs and agricultural history also allow us to explore some universal questions in ecology, such as how dispersal and environment limit species distributions.

Temperate deciduous forests across much of Europe and eastern North America reflect legacies of past land use, particularly in the diversity and composition of plant communities. Intense disturbances, such as clearing forests for agriculture, may cause persistent environmental changes that continue to shape vegetation patterns as landscapes recover. We assessed the long-term consequences of agriculture for environmental conditions in central New York forests, including tree community structure and composition, soil physical and chemical properties, and light availability. To isolate the effects of agriculture, we compared 20 adjacent pairs of forests that were never cleared for agriculture (primary forests) and forests that established 85-100 years ago on plowed fields (secondary forests). Tree communities in primary and secondary forests had similar stem density, though secondary forests had 14% greater basal area. Species composition differed dramatically between the two forest types, with primary forests dominated by Acer saccharum and Fagus grandifolia and secondary forests by Acer rubrum and Pinus strobus. Primary and secondary forests showed no consistent differences in soil physical properties or in the principal gradient of soil fertility associated with soil pH. Within stands, however, soil water content and pH were more variable in primary forests. Secondary forest soils had 15% less organic matter, 16% less total carbon, and 29% less extractable phosphorus in the top 10 cm than adjacent primary stands, though the ranges of the forest types mostly overlapped. Understory light availability in primary and secondary forests was similar. These results suggest that, within 100 years, post-agricultural stands have recovered conditions comparable to less disturbed forests in many attributes, including tree size and number, soil physical properties, soil chemical properties associated with pH, and understory light availability. The principal legacies of agriculture that remain in these forests are the reduced levels of soil organic matter, carbon, and phosphorus; the spatial homogenization of soil properties; and the altered species composition of the vegetation.

Assessing the relative roles of dispersal limitation and environmental effects in population dynamics and community assembly is fundamental to understanding patterns of species distribution and diversity. In forests growing on abandoned agricultural lands, both legacies of vegetation disturbance and changes in the abiotic environment shape the diversity and composition of recovering communities. Here I specify how interactions among historical, environmental, and biological factors influence species distributions, focusing on three fern species with contrasting distributions across forests of different history in central New York, USA: Dryopteris carthusiana, Dryopteris intermedia, and Polystichum acrostichoides. Using population surveys, spore-trap and spore-bank studies, and a three-year field experiment, I compare demographic rates among species and between forest types to determine which life history stages limit colonization and which traits explain species distributions. Adult plants of all three species were larger and more likely to produce spores in post-agricultural forests than in adjacent, uncleared stands. Though lower population densities led to fewer spores in post-agricultural soils, spore availability still exceeded recruitment by four to five orders of magnitude. Sowing additional spores had relatively little effect, while microhabitat conditions had the greatest impact on establishment rates. Given similar microsites, the two forest types had equal rates of establishment, but some forest-floor features preferentially occupied by juvenile plants were less frequent in post-agricultural stands. The availability of suitable sites for establishment, created by small-scale heterogeneity on forest floors, thus limits both the growth of fern populations and the colonization of new habitats. In fact, reduced microtopographic variation in post-agricultural forests may represent a greater hindrance to plant establishment than changes in mean environmental conditions. Among the three fern species, establishment rates differed as species distributions would predict, with the strongest colonizer consistently having the highest rates and the slowest colonizer the lowest. Rather than random or trait-mediated dispersal, the different distributions of these species reflect life history traits that determine establishment rates and thus colonization ability. This case study demonstrates that ecological interactions based on the unique life histories of individual species can override dispersal in determining species distributions.

Aim Climate, topography and soils drive many patterns of plant distribution and abundance across landscapes, but current plant communities may also reflect a legacy of past disturbance such as agricultural land use. To assess the relative influences of environmental conditions and disturbance history on vegetation, it is important to understand how these forces interact. This study relates the geographical distribution of land uses to variation in topography and soils; evaluates the consequences of land-use decisions for current forests; and examines the effects of agricultural land use on the chemical properties of forest soils.Location Tompkins County occupies 1250 km 2 in central New York's Finger Lakes region. Like much of eastern North America, this area underwent forest clearance for agriculture during the 1800s and widespread field abandonment and forest recovery during the 1900s. The current landscape consists of a patchwork of forests that were never cleared, forests that developed on old fields and active agricultural lands.Methods We investigated relationships among topography, soils and land-use decisions by gathering information about land-use history, slope, aspect, elevation, soil lime content, soil drainage and accessibility in a geographic information system (GIS). To assess the effects of agriculture on forest soil chemistry, we measured pH, organic matter content and extractable nutrient concentrations in field-collected soil samples from 47 post-agricultural and uncleared forests.Results Steeper slopes, less accessible lands and lower-lime soils tended to remain forested, and farmers were more likely to abandon fields that were steeper, farther from roads, lower in lime and more poorly drained. Slope had by far the greatest impact on patterns of clearance and abandonment, and accessibility had a surprisingly strong influence on the distribution of land uses. The effects of other factors varied more, depending for example on location within the county. Current forest types differed accordingly in topography and soil attributes, particularly slope, but they also showed much overlap. Post-agricultural and uncleared forest soils had similar chemical properties. Forests on lands abandoned from agriculture 80-100 years before had slightly higher pH and nutrient concentrations than adjacent, uncleared forests, but these changes were small compared to environmental variation across the county.Main conclusions Despite differential use of lands according to their topography and soils, the substantial influence of accessibility and the relatively small scale of land-use decisions allowed for broad similarity among forest types. Thus, the topography and soil differences created by land-use decisions probably contribute little to landscape-level patterns of diversity. Subtle changes in forest Journal of Biogeography (J. Biogeogr.) (2005) 32, 439-452 ª 2005 Blackwell Publishing Ltd www.blackwellpublishing.com/jbi

Atmospheric deposition contributes a large fraction of the annual nitrogen (N) input to the basin of the Susquehanna River, a river that provides two-thirds of the annual N load to the Chesapeake Bay. Yet, there are few measurements of the retention of atmospheric N in the Upper Susquehanna's forested headwaters. We characterized the amount, form (nitrate, ammonium, and dissolved organic nitrogen), isotopic composition (d 15 N-and d 18 O-nitrate), and seasonality of stream N over 2 years for 7-13 catchments. We expected high rates of N retention and seasonal nitrate patterns typical of other seasonally snow-covered catchments: dormant season maxima and growing season minima. Coarse estimates of N export indicated high rates of inorganic N retention ([95%), yet streams had unexpected seasonal nitrate patterns, with summer peaks (14-96 lmol L -1 ), October crashes (\1 lmol L -1 ), and modest rebounds during the dormant season (\1-20 lmol L -1 ). Stream d 18 O-nitrate values indicated microbial nitrification as the primary source of stream nitrate, although snowmelt or other atmospheric source contributed up to 47% of stream nitrate in some March samples. The autumn nitrate crash coincided with leaffall, likely due to in-stream heterotrophic uptake of N. Hypothesized sources of the summer nitrate peaks include: delayed release of nitrate previously flushed to groundwater, weathering of geologic N, and summer increases in net nitrate production. Measurements of shale d 15 N and soil-, well-, and streamwater nitrate within one catchment point toward a summer increase in soil net nitrification as the driver of this pattern. Rather than seasonal plant demand, processes governing the seasonal production, retention, and transport of nitrate in soils may drive nitrate seasonality in this and many other systems.

Summary1. Dispersal ability can influence the importance of dispersal relative to other processes organizing metacommunities, such as species sorting among habitats along environmental gradients. 2. We compare plants with different dispersal modes and habitat affinities, evaluating the roles of environmental and spatial controls on plant community composition in 128 wetlands within 10 km 2 of old-growth maple-beech forest in southern Que´bec, Canada.3. We address two hypotheses. First, we ask whether species with short-distance dispersal mechanisms are more dispersal-limited than species with adaptations for long-distance dispersal. Second, because wetland habitats are more fragmented than upland habitats in this forested area, we test the hypothesis that wetland species are more dispersal-limited than upland species growing in the same wetlands (e.g. on hummocks within a swamp). 4. Variation partitioning based on constrained ordinations showed that the distributions of species with short-distance dispersal mechanisms related more strongly to spatial factors than the distributions of long-distance dispersers, supporting the interpretation that these species are more dispersallimited. The distributions of short-distance dispersers also showed finer-scale spatial patterns than the distributions of long-distance dispersers. 5. Distributions of wetland species related more strongly to environmental conditions than the distributions of upland species growing in the same wetlands, suggesting that wetland species are actually less dispersal-limited than upland species. Wetland and upland species had similar patterns of spatial variation in community composition. 6. Synthesis. The processes of dispersal limitation and species sorting along environmental gradients have differential importance to plants that grow within the same communities, but differ in dispersal ability and habitat affinity. This result emphasizes the impact of dispersal ability on the organization of metacommunities.