Barriers to Forest Regeneration in an Abandoned Pasture in Puerto Rico

Sources of forest regeneration (soil seed bank, seed rain) and barriers to seedling establishment were examined in a recently abandoned pasture in eastern Puerto Rico. Few woody species were found in the soil seed bank or in the seed rain. The number of seeds and species in the seed rain and soil seed bank declined with distance from the adjacent secondary forest. Nine species naturally dispersed and colonized plots during the study, with the wind‐dispersed tree Tabebuia heterophylla being the predominant colonizer (91% of all seedlings). Barriers to seedling establishment were determined using a blocked field experiment with eleven woody species representative of a variety of life histories. Each species was planted under the pasture vegetation (control) or in areas where all vegetation was removed (removal). Germination was enhanced for four species in the control treatment, five species were not affected, and two species did not germinate under either treatment. Survival to 6 months was higher in the removal treatment for two species. Seedling biomass was greater in the removal treatment at 12 months for one species. Seed mass was a good predictor of germination success and final shoot biomass, but not survival. This study demonstrates that seeding recently abandoned pastures with a mix of known pioneer species may accelerate the rate of secondary succession, but some species will have to be planted in later successional stages in order to overcome strong barriers to establishment.     

Mechanical Land Clearing to Promote Establishment of Coastal Sandplain Grassland and Shrubland Communities

The decline in grasslands and other species‐rich early successional habitats on the coastal sandplains of the northeastern United States has spurred management to increase the area of these declining plant communities. We mechanically removed overstory oak and applied seed from a nearby sandplain grassland on the island of Martha’s Vineyard, Massachusetts, to evaluate this technique for creating an open oak community able to support sandplain herbaceous species. We compared vegetation structure and composition before and after clearing in an area of total tree removal (clearcutting), an area where 85% of tree basal area was removed (savanna cutting), and in adjacent coastal oak forest. Plant responses to clearcutting and savanna cutting were similar. Sandplain herbs colonized at high frequencies after seeding and increased herbaceous cover from less than 7% before clearing to 22–38% three growing seasons later. Pennsylvania sedge (Carex pensylvanica) increased in cover approximately 6‐fold, accounting for 84–90% of the increased herbaceous cover. Other native ruderals and exotic herbs reached 2 and less than or equal to 1%, cover, respectively, after three years. Species richness across cleared treatments increased from 30 to 79 species. All forest species were retained. Forest shrubs and trees initially declined from their dominant cover but rebounded after three years. Tree clearing plus seeding appeared to be a viable management practice for increasing cover of herbaceous sandplain species while causing minimal increases in exotic herbaceous cover. The long‐term persistence of sandplain herbs may require periodic disturbances that limit woody regrowth.     

Causes of plant community divergence in the early stages of volcanic succession

Question: How do temporal changes in plant communities occur after volcanic eruptions? What characteristics determine successional divergence or convergence? Location: The summit area of Mount Usu, northern Japan, completely destroyed by 1‐3 m of thick ash and pumice during the 1977‐1978 eruptions. Habitats were classified into three types: gullies where the pre‐eruption topsoil was exposed due to the erosion of tephra (EG), gullies covered with tephra (CG), and outside of gullies covered with thick tephra (OG). Methods: Plant community structure was monitored for 15 years from 1983 to 1997 in 14 2 m × 5 m permanent plots. The data were summarized by species diversity, life form, and the detrended correspondence analysis. Results: The common species were perennial herbaceous plants, but habitat preferences differed between species. Seed bank species, including a nitrogen fixer Trifolium repens, were dominant in EG, and excluded the establishment of the later colonists. Pioneer trees slowly increased in cover. The detrended correspondence analysis indicated that species composition in the earlier stages did not differ greatly between plots. Thereafter, three patterns of temporal community changes were observed: seed bank species persisted in EG, and in OG and CG forest development proceeded or community structure did not change greatly. Conclusion: Pre‐eruption topsoil contributed to revegetation by the supply of seed bank and nutrients in the earliest stages, but resulted in the delay of forest development due to the persistence of seed bank species. Plant community divergence was driven by the persistence of earlier colonists.     

Experimental assessment of initial revegetation on abandoned paths in temperate deciduous forest

Question: Do soil treatments and addition of seed facilitate rapid vegetation restoration on forest paths excluded from trampling? Location: Six mesophilic mixed deciduous Querco‐Fagetea forests in Flanders, northern Belgium. Methods: Enclosures on paths were excluded from trampling by fencing. In a full factorial design, plots were subjected to seeding, soil scarification, addition of organic material and inoculum. The following two years, seedling establishment and growth were sampled during spring and summer. Soil treatments and seeding effect was tested and seedling occurrence was analysed in relation to species' origin. Results: Spontaneous re vegetation was significant in all plots since fencing. Throughout the observation period seedling cover and height continued to increase. Seeding had an overall effect on seedling density, cover and height. Some soil treatment interactions significantly enhance re vegetation, although each individual soil treatment had no significant effect. Regardless of the seeded individuals, invading species mostly originated from the surrounding area and the seed bank. Conclusion: The preliminary results of this experiment imply that seeding is the only treatment which has a positive effect on re vegetation success in all circumstances, provided that the exclusion from trampling is effective.     

Importance of Seed and Microsite Limitation: Native Wildflower Establishment in Non‐native Pasture

Barriers to establishing native plant communities on former pasture include dominance by a single planted species, hydrologic and edaphic alteration, and native species propagule limitation. Establishment may be dispersal‐limited (propagules do not arrive at the site), microsite‐limited (areas suitable for seedling emergence and survival do not exist), or both. Successful restoration strategies hinge on identifying and addressing critical limitations. We examined seed and microsite limitation to establishment of a native wildflower (Coreopsis lanceolata ) in a former pasture dominated by Paspalum notatum (bahiagrass). We determined the relative and interactive effects of microsite (irrigation and disturbance) and seed limitation on C. lanceolata establishment. We tested (1) irrigation (none, pre‐seeding, and pre‐ and post‐seeding), (2) disturbance (none, sethoxydim, glyphosate, and topsoil removal), and (3) C. lanceolata seeding rate (three seeding densities). Applying glyphosate before seeding increased C. lanceolata establishment compared to other disturbance treatments. Ultimately, C. lanceolata establishment was not affected by irrigation. Coreopsis lanceolata establishment was limited when seeded at 100 live seeds/m² but not at 600 or 1100 live seeds/m². Seed and microsite availability interactively affected C. lanceolata establishment, in that microsite limitation was biologically relevant only when a minimum number of seeds were present. In practice, both seed and microsite requirements must be met for successful establishment, and increasing the availability of seeds or microsites does not compensate for limitations of the other. Here, it is the relative importance of seed and microsite limitations that drives plant establishment; these limitations do not represent a simple dichotomy.     

Roles of non‐native species in large‐scale regeneration of moist tropical forests on anthropogenic grassland

This paper presents a new synthesis of the role of native and non‐native species in diverse pathways and processes that influence forest regeneration on anthropogenic grassland in the moist tropics. Because of altered species composition, abiotic conditions and landscape habitat mosaics, together with human interventions, these successional pathways differ from those seen in pre‐clearing forests. However, representation of different functional life forms of plant (tree, vine, grass, herb and fern) and animal (frugivorous seed disperser, granivorous seed predator, seedling herbivore and carnivore) shows consistent global variation among areas of pasture, intact forest, and post‐grassland regrowth. Biotic webs of interaction involve complex indirect influences and feedbacks, which can account for wide observed variation in regeneration trajectories over time. Important processes include: limitation of tree establishment by dense grasses; recruitment and growth of pioneer pasture trees (shading grasses and facilitating bird‐assisted seed dispersal); and smothering of trees by vines. In these interactions, species’ functional roles are more important than their biogeographic origins. Case studies in eastern Australia show native rain forest plant species diversity in all life forms increasing over time when pioneer trees are non‐native (e.g., Cinnamomum camphora, Solanum mauritianum), concurrent with decreased grass and fern cover and increased abundance of trees and vine tangles. The global literature shows both native and non‐native species facilitating and inhibiting regeneration. However conservation goals are often targeted at removing non‐native species. Achieving large‐scale tropical forest restoration will require increased recognition of their multiple roles, and compromises about allocating resources to their removal.     

Managing open habitats by wild ungulate browsing and grazing: A case‐study in North‐Eastern Germany

Question: Can wild ungulates efficiently maintain and restore open habitats? Location: Brandenburg, NE Germany. Methods: The effect of wild ungulate grazing and browsing was studied in three successional stages: (1) Corynephorus canescens‐dominated grassland; (2) ruderal tall forb vegetation dominated by Tanacetum vulgare; and (3) Pinus sylvestris‐pioneer forest. The study was conducted over 3 yr. In each successional stage, six paired 4 m²‐monitoring plots of permanently grazed versus ungrazed plots were arranged in three random blocks. Removal of grazing was introduced de novo for the study. In each plot, percentage cover of each plant and lichen species and total cover of woody plants was recorded. Results: Wild ungulates considerably affected successional pathways and species composition in open habitats but this influence became evident in alteration of abundances of only a few species. Grazing effects differed considerably between successional stages: species richness was higher in grazed versus ungrazed ruderal and pioneer forest plots, but not in the Corynephorus sites. Herbivory affected woody plant cover only in the Pioneer forest sites. Although the study period was too short to observe drastic changes in species richness and woody plant cover, notable changes in species composition were still detected in all successional stages. Conclusion: Wild ungulate browsing is a useful tool to inhibit encroachment of woody vegetation and to conserve a species‐rich, open landscape.     

Seed to seedling transitions in successional habitats across a tropical landscape

Recognition that tree recruitment depends on the balance between seed arrival and seedling survival has led to a surge of interest in seed‐dispersal limitation and seedling‐establishment limitation in primary forests. Virtually unaddressed are comparisons of this balance in mature and early successional habitats. We assessed seed rain and seedling recruitment dynamics of tree species in primary forest, secondary forest and pasture released from grazing in a tropical agricultural landscape. Seed to seedling ratios (seed effectiveness; Φ_i) for 43 species in southern Mexico determined differences in the extent to which seeds produced seedlings by habitat, life history, and dispersal mode. Reproductive potential as estimated by the transition from seed rain to seedling recruitment, differed by habitats, and varied dramatically by life history and dispersal mode. Expected recruit densities (E_it) were higher for animal‐dispersed than wind‐dispersed species, and for non‐pioneer than pioneer species. Non‐pioneers and animal‐dispersed species had higher expected relative recruit abundance (ε_it) in primary forest (median of 4 seeds recruit^?1) whereas in secondary forest wind‐dispersed pioneers had the highest expected relative recruit abundance (median of 16 seeds per recruit). In pastures, wind‐dispersed pioneer species were most successful with many more seeds per recruit (median of 291) than both forest habitats. Seeds per recruit (Φ_i) appeared to decrease with increase in seed mass for 43 species for which data were available (r = –0.55, P < 0.001). This was associated with a negative correlation of Φ_i with seed size in primary forest (r = –0.50, P = 0.08 for 13 species); Φ_i was not correlated with seed size in secondary forest (n = 16) or pasture (n = 14). Metrics of seeds per recruit, expected recruit density and expected relative recruit abundance dramatically illustrate differences in barriers to recruitment in successional habitats.     

Non-random recruitment of invader species in experimental grasslands

To assess potential effects of seed limitation, characteristics of invader species and characteristics of established plant communities on recruitment success, we conducted a split-plot experiment factorially combining three weeding treatments corresponding to increasing successional age (regular weeding相似文献   

Effects of Small Rodent and Large Mammal Exclusion on Seedling Recruitment in Costa Rica1

We examined whether the experimental exclusion of large mammalian and small rodent seed predators had differing effects on seedling recruitment under natural seed rain conditions. In both primary and late‐successional secondary forested areas, exclosure experiments using natural seed densities were designed to assess seedling recruitment. To assess the differences in seedling recruitment, we monitored three exclosure treatments (1.2 m radius/1.5 m height) in two forest types (primary vs. late‐successional secondary forest): (1) fenced exclosures that excluded large mammals; (2) fenced exclosures that excluded both large and small mammals; and (3) open controls. Within each exclosure treatment, we marked and identified all seedlings at the beginning of the experiment (February 2001), followed the marked seedlings' fate for a year, and then marked and identified all new seedlings after a year. Two preliminary findings were generated from these data: for some tree species, small rodents and large mammals have differential effects on seedling recruitment, and the effect of excluding mammals did not differ with habitat type (primary vs. late‐successional secondary forest). These preliminary results highlight the need to examine further how the effects of small rodent and large mammal exclusion may affect species‐specific seed predation and seedling recruitment in a variety of habitat/land use types (e.g., primary forest, late‐successional forest, and early‐successional forest).     

Restoration of Tropical Moist Forests on Bauxite-Mined Lands in the Brazilian Amazon

We evaluated forest structure and composition in 9- to 13-year-old stands established on a bauxite-mined site at Trombetas (Pará), Brazil, using four different reforestation techniques following initial site preparation and topsoil replacement. These techniques included reliance on natural forest regeneration, mixed commercial species plantings of mostly exotic timber trees, direct seeding with mostly native early successional tree species, and mixed native species plantings of more than 70 tree species (the current operational restoration treatment at this site). Replicated fixed-radius plots in each treatment and in undisturbed primary forest were used to quantify the canopy and understory structure and the abundance and diversity of all vascular plant species. Treatment comparisons considered regeneration density, species richness and diversity for all floristic categories, and, for trees and shrubs, the relative contribution of initial planting and subsequent regeneration from soil seed banks and seed inputs from nearby primary forests. With the possible exception of the stands of mixed commercial species, which were superior to all others in terms of tree basal-area development but relatively poor in species richness, all treatments were structurally and floristically diverse, with a high probability of long-term restoration success. Of these, the mixed native species plantings appeared to be at least risk of arrested succession due to the dominance of a broader range of tree species of different successional stages or expected life spans. In all treatments, several locally important families of primary forest trees (Annonaceae, Chrysobalanaceae, Lauraceae, Palmae and Sapotaceae) were markedly underrepresented due to a combination of poor survival of initial plantings and limitations on seed dispersal from the surrounding primary forest.     

Supplementing Seed Banks to Rehabilitate Disturbed Mojave Desert Shrublands: Where Do All the Seeds Go?

Revegetation of degraded arid lands often involves supplementing impoverished seed banks and improving the seedbed, yet these approaches frequently fail. To understand these failures, we tracked the fates of seeds for six shrub species that were broadcast across two contrasting surface disturbances common to the Mojave Desert—sites compacted by concentrated vehicle use and trenched sites where topsoil and subsurface soils were mixed. We evaluated seedbed treatments that enhance soil‐seed contact (tackifier) and create surface roughness while reducing soil bulk density (harrowing). We also explored whether seed harvesting by granivores and seedling suppression by non‐native annuals influence the success of broadcast seeding in revegetating degraded shrublands. Ten weeks after treatments, seeds readily moved off of experimental plots in untreated compacted sites, but seed movements were reduced 32% by tackifier and 55% through harrowing. Harrowing promoted seedling emergence in compacted sites, particularly for the early‐colonizing species Encelia farinosa, but tackifier was largely ineffective. The inherent surface roughness of trenched sites retained three times the number of seeds than compacted sites, but soil mixing during trench development likely altered the suitability of the seedbed thus resulting in poor seedling emergence. Non‐native annuals had little influence on seed fates during our study. In contrast, the prevalence of harvester ants increased seed removal on compacted sites, whereas rodent activity influenced removal on trenched sites. Future success of broadcast seeding in arid lands depends on evaluating disturbance characteristics prior to seeding and selecting appropriate species and seasons for application.     

Establishment of Picea abies seedlings in a central European mountain grassland: an experimental study

Abstract. Secondary meadows in the Kru?né hory mountains, central Europe, are not colonized by woody species, although they have been abandoned for half a century and are surrounded by Picea abies (Norway spruce) forests. The causes of inhibition of establishment of Norway spruce seedlings in the meadows were tested experimentally. The experiment was started in a masting year to ensure sufficient seed input. Four treatments (vegetation cut; vegetation cut and litter removed; all above-ground biomass and topsoil removed; control) were combined in a factorial design. The effect of browsing, mainly by deer, was assessed by fencing half of the experimental plots. Seedling establishment differed among treatments (all pair-wise differences were significant) but was independent of fencing. In contrast, seedling survival was influenced both by the treatment and fencing. At the end of the second year, only the plots with all biomass and organic topsoil removed supported viable populations of spruce seedlings. Dense herb cover and a thick layer of slowly decaying litter are considered the main factors inhibiting the establishment of woody species.     

The Effect of Spatial Structure of Pasture Tree Cover on Avian Frugivores in Eastern Amazonia

The movement of frugivores between remnant forests and successional areas is vital for tropical forest tree species to colonize successional habitats. The response of these species to the spatial structure of pasture tree cover is largely unknown. We studied avian frugivores that were found in primary forest edges and large pastures in eastern Amazonia, Brazil. We determined how the small‐scale spatial structure of pasture trees at forest edges affects five response variables: bird presence, visitation rate, duration of visit, species richness, and an index accounting for species’ level of frugivory and abundance in forests. We used hierarchical linear models to estimate the effect of four predictor variables on response variables: (1) clustering of pasture trees; (2) percent canopy cover of pasture trees; (3) distance of pasture tree to forest edge; and (4) tree crown area. The study species, many of which are widely distributed in the Neotropics, were generally insensitive to percent cover and clustering of trees. Frugivore visitation to individual trees remained constant as cover increased. Visitation was positively correlated with focal tree distance to forest edge and crown area. The positive relationship between distance and visitation rates may be due to the increased abundance of some resource further from forests. If pastures were abandoned the distance from forest edges would not likely limit frugivore visitation and seed deposition under large pasture trees in our study (i.e., up to 200 m distant).     

Traits and growth of liana regeneration in primary and secondary forests of Central Amazonia

Question: Do traits of liana regeneration differ among secondary forest types of varying land‐use history and primary forest? Location: Eighty kilometers north of Manaus, Brazil. Methods: We compared plant functional traits and growth rates of liana regeneration (<1.7‐m length) among two secondary forest types and primary forest. Secondary forest types were: Vismia (on land formerly clear‐cut, used for pasture and intensively burned) and Cecropia (no pasture usage or intensive fires after clear‐cut). Results: A principal components analysis indicated that most of the primary forest species exhibited a similar habit and were characterized by short shoots and small, round leaves with low specific leaf area, whereas secondary forest species had a broad range of trait values. At the plot level, primary and secondary forest communities were separated mainly by plant length and leaf size. Plant size varied more within secondary than within primary forest plots. The two secondary forest types could not be separated based on the traits of liana regeneration. Relative growth rate (RGR) did not correlate significantly with any measured plant trait, except for a negative relation to initial length. RGR increased with decreasing canopy cover and was highest in Vismia forest plots. Conclusion: Plant functional traits of liana regeneration were more similar in the primary forest and differed substantially from secondary forests, yet canopy cover only partly explained the observed differences.     

Testing Broadcast Seeding Methods to Restore Urban Forests in the Presence of Seed Predators

Forest restoration in urban areas often occurs in isolation from remnant forest, limiting the chances for recolonization by native species. Plants with bird‐dispersed seeds can be particularly vulnerable to dispersal limitation and regeneration can be further impeded by non‐native seed predators. We used a factorial experiment to investigate broadcast seeding as a method to reintroduce trees with large seeds and fleshy fruits into early successional forests. We assessed rates of seed and fruit loss, germination and seedling establishment in three seed treatments: (1) caging to exclude introduced mammalian seed predators; (2) removal of fleshy fruit pericarp; and (3) placing seeds in nutritionally enriched clay balls. Across all species (Beilschmiedia tawa, Elaeocarpus dentatus, and Litsea calicaris) seeds and fruits accessible to mammalian predators suffered significantly greater loss (58%) than those protected by cages (4%). However, seed and fruit loss in the presence of predators was reduced to only 35% across all species by the treatment combining the removal of fruit flesh and clay ball application to seeds. Establishment of B. tawa seedlings after 1 year was significantly enhanced by the clay ball treatment (12% of seeds sown vs. 6% without clay balls). Very low establishment rates were recorded for E. dentatus and L. calicaris. Broadcast seeding was found to be a viable method of improving regeneration of large‐seeded late successional trees and may be a cost‐effective alternative to planting saplings. Seedling establishment can be improved with fruit flesh removal and clay ball treatments, especially in the presence of mammalian seed predators.     

Soil Seed Bank as an Input of Seed Source in Revegetation of Lead / Zinc Mine Tailings

The goal of the present study was to assess a soil seed bank as an input seed source for revegetating lead/zinc (Pb/Zn) mine tailings. The seed bank source was abandoned farmland, whose top 10‐cm layer of topsoil contained 6,850 ± 377 seeds/m² from 41 species. The seeds in the soil were principally distributed in the upper 0–2 cm, which held 75.8% of total seeds and 92.7% of species composition. The top 2‐cm layer of topsoil may be sufficient to serve the purpose of providing a seed source for revegetation on derelict lands, including mined lands. Four different thicknesses of topsoil (1, 2, 4, and 8 cm, redistributed from the total 0–10‐cm layer from the farmland) were field‐tested on the Pb/Zn mine tailings. There was no significant difference in seedling density among the 4 thickness treatments. Many seeds in the treatments with more than 1‐cm of topsoil were unable to emerge from the deeper layer. Seedlings in plots with topsoil of 1‐, 2‐ and 4‐cm failed to establish within 1 year due to the extremely high acidity (pH 2.39 to 2.76). A shallow layer of topsoil cannot neutralize the sulfuric acid generated from oxidation of pyrites in the tailings. For establishment of seedlings on metalliferous lands, an insulating layer such as subsoil, building rubble, or domestic refuse is necessary before covering with valuable topsoil. The woody legume Leucaena leucocephala grown on the tailings with a topsoil cover of 8‐cm was the most dominant species. Lead was accumulated in root, branch, stem bark, and xylem, which accounted for more than 80% of the total metal concentration in the plant. This portion of Pb will reside in the plant for a long period, while the smaller portion of Pb in the leaf (about 15%) could be returned to the environment as litter during growth. Woody plants may have an advantage in metal‐phyto‐remediation over herbaceous plants.     

Soil seed banks of fringing salt lake vegetation in arid Western Australia – density,composition and implications for postmine restoration using topsoil

Although studies of seed banks in arid ecosystems are commonplace, they are lacking for the large arid zone of Western Australia. Across the six major plant communities fringing a large salt lake within this zone, topsoil (0–5 cm depth) was collected from 12 to 36 sites per community. Samples were dried, spread out on a bed of vermiculite in seedling trays and placed in a well‐watered glasshouse to determine the readily germinable component of the soil seed bank. Subsamples of topsoil were treated with smoke water, hot water or flooding to help determine seed bank of species with dormancy mechanisms. As with other studies of arid seed banks, large numbers of grasses and forbs emerged from the topsoil, with relatively small numbers of woody perennial species and hummock grasses (Triodia spp.) present, even in communities where such species were dominant. There were, however, a few exceptions where a reasonable density of dominant trees/shrub seed was present in topsoil. Soil treatment generally had limited effect on composition and density of emergent seedlings. Although floristic similarity between soil seed banks and corresponding above‐ground vegetation was modest, there were clear differences in soil seed bank composition between communities. The implications of the results for using topsoils to restore landforms of the study area after mining or other disturbance are discussed.     

Bird Perches Increase Forest Seeds on Puerto Rican Landslides

Abstract Landslides result in the loss of vertical vegetative structure, soil nutrients, and the soil seed bank. These losses impede timely recovery of tropical forest communities. In this study we added bird perches to six Puerto Rican landslides with three types of surfaces (bare, climbing fern, grass) in an effort to facilitate inputs of forest seeds through bird dispersal and to accelerate plant succession. Numbers of bird‐dispersed forest seeds were significantly higher in plots beneath introduced perches than in control plots. Perches did not increase forest seedling densities compared with control plots. Seven different species of birds were observed on introduced perches. Because 99% of the seed inputs to controls and perch plots in the six landslides were wind‐dispersed seeds (mostly graminoids), perches can improve landslide restoration if woody plants establish and shade out the dominant graminoid and climbing fern ground cover. Although increasing seed inputs from forest species is a critical step in accelerating revegetation of landslides, we suggest that supplemental restoration techniques be applied in addition to bird perches to promote forest recovery.     

Seed Bank Effects on Recovery after Disturbance in Reconstructed Tallgrass Prairies

Litter‐removing disturbances such as fire in grasslands temporarily increase available resources for plants, opening a window of opportunity for new establishment as communities recover. At this time, new individuals or species could be added to the community as a result of germination from the local seed bank. In reconstructed grasslands this may be problematic, as the seed bank may contain a suite of undesired species reflective of prior and surrounding land uses. In two, 25‐year‐old, low‐diversity reconstructed grasslands, we tested the effect of local seed bank establishment following litter‐removing disturbance using seedling removal plots (1 m²) and plots where natural seedling establishment was allowed. Following disturbance, the vegetation was either left intact or hayed to enhance seedling establishment (a common practice following inter‐seeding efforts). Although the seed bank and seedling community were dominated by resident grasses (Andropogon gerardii and Poa pratensis), recruitment from the seed bank increased species richness and reduced evenness through the addition of forb species (including Cirsium arvense) in one of the study sites. Haying temporarily altered the abundances of the dominant grasses, but did not consistently affect seedling recruitment. Disturbances that facilitate seed bank recruitment may promote establishment of undesired species within reconstructed grassland communities, and we need to take steps to better manage the contributions into and recruitment from the seed bank to reconstruct sustainable grasslands.