Initial Plant Community Research Articles

Local environmental context drives heterogeneity of early succession dynamics in alpine glacier forefields

Abstract. Glacier forefields have long provided ecologists with a model to study patterns of plant succession following glacier retreat. While plant-survey-based approaches applied along chronosequences provide invaluable information on plant communities, the “space-for-time” approach assumes environmental uniformity and equal ecological potential across sites and does not account for spatial variability in initial site conditions. Remote sensing provides a promising avenue for assessing plant colonization dynamics using a so-called “real-time” approach. Here, we combined 36 years of Landsat imagery with extensive field sampling along chronosequences of deglaciation for eight glacier forefields in the southwestern European Alps to investigate the heterogeneity of early plant succession dynamics. Based on the two complementary and independent approaches, we found strong variability in the time lag between deglaciation and colonization by plants and in subsequent growth rates and in the composition of early plant succession. All three parameters were highly dependent on the local environmental context, i.e., neighboring vegetation cover and energy availability linked to temperature and snowmelt gradients. Potential geomorphological disturbance did not emerge as a strong predictor of succession parameters, which is perhaps due to insufficient spatial resolution of predictor variables. Notably, the identity of pioneer plant species was highly variable, and initial plant community composition had a much stronger influence on plant assemblages than elapsed time since deglaciation. Overall, both approaches converged towards the conclusion that early plant succession is not stochastic as previous authors have suggested but rather determined by local ecological context. We discuss the importance of scale in deciphering the complexity of plant succession in glacier forefields and provide recommendations for improving botanical field surveys and using Landsat time series in glacier forefield systems. Our work demonstrates complementarity between remote sensing and field-based approaches for both understanding and predicting future patterns of plant succession in glacier forefields.

Vertical Columns with Sustainable Green Cover: Meadow Plants in Urban Design.

Unique vertical column structures were constructed for the greening of a structure at the Botanical Garden of Vilnius University, in which a plant cover was formed using the turf rolls of semi-natural meadows that were wrapped on 197 columns, with each column consisting of three equal segments. By evaluating the species composition and the abundance of vegetation in the segments of the columns, we studied how this natural cover changes and what its survival potential is. During the five years of observation, 97 plant species were determined in total. Over time, the initial plant species of fertile soils were mostly replaced by ruderal, nitrophilous, and pioneer plants. Out of the 58 original species, 18 disappeared, while 39 new ones emerged. In the vegetation cover on the north exposition of the building, the original species composition declined faster. The most persistent species were ruderal short-lived Conyza canadensis, Melilotus albus, and Urtica dioica, and long-lived Elytrigia repens. As for vegetation classes, the initial plant communities of the Molinio-Arrhenatheretea elatioris vegetation class were partially replaced by the plant communities of the Koelerio-Corynephoretea canescentis and Artemisietea vulgaris classes; however, unformed plant communities finally became prevalent. All directions, including the north, east, south and west, were equally dominated by semi-shade- and semi-light-loving plant species, together with a less abundant representation of light-loving species. Meanwhile, an unexpected establishment of the light-loving annual Arabidopsis arenosa was observed on the least illuminated north exposition. Likewise, the perennial Festuca pratensis, which is particularly resistant to wintering, emerged and spread on all expositions. The vegetation in the vertical columns was dynamic, and the initial plant species significantly diminished in the five years; however, as new species took place, the columns remained sufficiently covered with a green carpet of plants. This study reveals the benefits of using semi-natural meadow turfs in vertical greening of buildings in the harsh climate of a 5b hardiness zone, which is accompanied by distressing climatic fluctuations during the vegetation season.

Initial plant community responses to hardwood control treatments in restoration of remnant longleaf pine (Pinus palustris) woodlands

Changes in land use over the past century have contributed to substantial losses of longleaf pine (Pinus palustris) woodlands in the southeastern USA and replacement with higher density, mixed pine and hardwood stands that suppress understory development and limit application of prescribed fire. To increase understanding of limiting factors and identify potential approaches for restoring longleaf pine woodlands, we studied initial availability of light and soil water and 2-year (2018–2019) plant community responses after controlling overstory hardwoods in five remnant longleaf pine woodlands having no evidence of previous agriculture at the Savannah River Site near Aiken, SC, USA. Seven hardwood control treatments and a non-treated check were compared in a randomized compete block experiment: cutting, cutting + shredding of logging residues, stem injection with imazapyr herbicide, cutting + basal spray with imazapyr herbicide, cutting + basal spray with triclopyr herbicide, cutting + directed foliar spray with a mixture of glyphosate and imazapyr herbicides, and cutting + broadcast foliar spray with the same herbicide mixture. In the year prior to hardwood cutting (2016), understory light availability averaged 23% of full sun. Throughout the year of treatment installation (2017), average soil water content (SWC) was below permanent wilting point (5.5% SWC) where overstory hardwoods were retained; whereas, it was above wilting point (7.5%) where they had been cut. Thus, combined effects of shade and root competition from overstory hardwoods probably limited cover of herbaceous species in the non-treated check. In the 2 years following treatment, hardwood survival averaged > 99%, 62%, 42% and < 1% for the non-treated check, cutting, shredding, and herbicide treatments, respectively. Relative to the non-treated check, herbaceous species richness was increased by shredding, stem injection, basal spray, or directed foliar treatments; herbaceous cover was increased by basal spray or directed foliar treatments; and woody cover was decreased by directed or broadcast foliar treatments. The directed foliar spray was the most effective treatment for achieving many of the desired understory characteristics of a longleaf pine woodland, including a diverse understory dominated by herbaceous vegetation capable of supporting periodic prescribed fires. Although the imazapyr basal spray treatment increased cover of remnant woodland indicator species, none of the treatments increased richness of this species group, suggesting that restoration of native species composition will be expedited by enrichment seedings or plantings.

Richness of Cerrado Woody Species Engaged in Ecological Restoration in the Brazilian Federal District

Abstract This study has evaluated the richness of Cerrado woody species engaged in ecological restoration in the Brazilian Federal District (BFD). A survey gathered information on plant species traded by local nurseries, species recommended in restoration plans (PRADs), species effectively introduced in areas under restoration, and species present in preserved fragments of Cerrado. Results summed 566 Cerrado woody species from 80 botanical families of which 171 species were traded by local nurseries, 277 were recommended in PRADs, 190 were effectively used in restoration projects, and 434 species were sampled in fragments of native Cerrado. We found low similarity between species composition available in nurseries, recommended in PRADs, used in restoration projects and present in preserved fragments of native Cerrado. Such results indicate a poor connection between steps related to the selection of native woody species that make up initial plant communities on sites under ecological restoration.

On the paleobotanical characteristic of the mospino formation (middle carboniferous, Donets Basin)

Formulation of the problem. Deposits of the Mospino Formation (Upper Bashkirian) are poorly characterized by plants remains. Any new data supplementing the paleobotanical characterization of the Carboniferous deposits of the Donets Basin are, undoubtedly, of interest. The aims of the article are to find out the systematic composition of plant remains from two localities and to determine the conditions of growth and burial of plants. The history of the study of Carboniferous flora of the Donets Basin dates back almost two centuries. During this long period, a rich fossil macroflora was investigated from the Carboniferous sediments. Material and research methods. This article is based on the results of a study of fossil flora from two sites – Makedonovka and Volnuhino (Lutuginsky district of the Luhansk region). Statement of the main material. Of the sediments with the coal layer g12 (Makedonovka), 32 taxa were identified: 20 species and 12 forms defined in the open nomenclature. The composition of the plant complex is as follows: Bothrodendron minutifolium, Cyperites bicarinatus, Halonia sp., Lepidodendron lycopodioides, Lepidophloios laricinus, Lepidostrobophyllum sp., Syringodendron sp. 2, Stigmaria ficoides, Asterophyllites grandis, Asterophyllites longifolius, Calamites carinatus, Calamites undulatus, Calamites cistii, Calamites sp., Pinnularia capillacea, Sphenophyllum cuneifolium, Sphenophyllum sp., Alethopteris sp. 2, Dictyoxylon sp., Eusphenopteris cf. obtusiloba, Eusphenopteris sp., Karinopteris acuta, Karinopteris sp., Neuralethopteris rectinervis, Neuralethopteris schlehanii, Lyginopteris hoeninghausi, Paripteris gigantea, Trigonocarpus parkinsonii, ?Trigonocarpus sp., Cordaicarpus cordai, Cordaites principalis, Cordaites sp. From the roof of the coal layer g3 (Volnuhino), 36 plant taxa were determined: 21 species and 15 forms defined in the open nomenclature. Among them: Asolanus camptotaenia, Bothrodendron minutifolium, Cyperites bicarinatus, Knorria sp., Lepidodendron obovatum, Lepidodendron aculeatum, Lepidophloios laricinus, Lepidostrobophyllum majus, Syringodendron sp. 1, Stigmaria ficoides, Asterophyllites charaeformis, Asterophyllites grandis, Calamites carinatus, Calamites undulatus, Calamites cf. sachsei, Calamites sp., Calamostachys sp., Sphenophyllum cuneifolium, Alethopteris davreuxi, Alethopteris sp.1, Aulacopteris sp., Cardioneura amadoca, Cyclopteris sp., Karinopteris acuta, Karinopteris beneckei, Karinopteris cf. dernoncourti, Mariopteris cf. nervosa, Mariopteris sp., Lyginopteris hoeninghausi, Neuropteris cf. obliqua, Neuropteris sp., Palmatopteris furcata, Tetragonocarpus palibinii, Artisia approximata, Cordaites sp., Samaropsis sp. A detailed morphological comparison of the following group of species is necessary: Cardioneura amadoca, Neuropteris bulupalganensis, Neurocardiopteris asiatica, as well as species of the genus Neurocardiopteris Lutz. It is possible that Neuropteris bulupalganensis and Neurocardiopteris asiatica, having close age and morphology, will turn out to be synonyms. Hygrophilic and meso-hygrophilic plant communities were the initial plant communities for Volnukhino phytoorictocenosis. The deposits of the Mospino Formation belong to the zone of joint distribution Neuralethopteris spp. – Lyginopteris hoeninghausii (NL) schemes by O.P. Fisunenko or macrofloristic subzone Alethopteris decurrens of the Lyginopteris hoeninghausi zone of the scheme of N.I. Boyarina. Conclusions. Fossil site Volnuhino is the richest locality of the flora remains of the Mospino Formation. The deposits are sediments of a freshwater lake, peat, lagoon, delta, and coastal accumulative plain. Volnuhino’s locality deserves to be granted the status of a conservation object – a paleontological geosite.

Experimental salt marsh islands: A model system for novel metacommunity experiments

Shallow tidal coasts are characterised by shifting tidal flats and emerging or eroding islands above the high tide line. Salt marsh vegetation colonising new habitats distant from existing marshes are an ideal model to investigate metacommunity theory. We installed a set of 12 experimental salt marsh islands made from metal cages on a tidal flat in the German Wadden Sea to study the assembly of salt marsh communities in a metacommunity context. Experimental plots at the same elevation were established within the adjacent salt marsh on the island of Spiekeroog. For both, experimental islands and salt marsh enclosed plots, the same three elevational levels were realised while creating bare patches open for colonisation and vegetated patches with a defined transplanted community. One year into the experiment, the bare islands were colonised by plant species with high fecundity although with a lower frequency compared to the salt marsh enclosed bare plots. Initial plant community variations due to species sorting along the inundation gradient were evident in the transplanted vegetation. Competitive exclusion was not observed and is only expected to unfold in the coming years. Our study highlights that spatially and temporally explicit metacommunity dynamics should be considered in salt marsh plant community assembly and disassembly.

Decreased cryogenic disturbance: one of the potential mechanisms behind the vegetation change in the Arctic

During the last few decades, the Arctic has experienced large-scale vegetation changes. Understanding the mechanisms behind this vegetation change is crucial for our ability to predict future changes. This study tested the hypothesis that decreased cryogenic disturbances cause vegetation change in patterned ground study fields (non-sorted circles) in Abisko, Sweden during the last few decades. The hypothesis was tested by surveying the composition of plant communities across a gradient in cryogenic disturbance and by reinvestigating plant communities previously surveyed in the 1980s to scrutinise how these communities changed in response to reduced cryogenic disturbance. Whereas the historical changes in species occurrence associated with decreased cryogenic disturbances were relatively consistent with the changes along the contemporary gradient of cryogenic disturbances, the species abundance revealed important transient changes highly dependent on the initial plant community composition. Our results suggest that altered cryogenic disturbances cause temporal changes in vegetation dynamics, but the net effects on vegetation communities depend on the composition of initial plant species.

Relative importance of abiotic, biotic, and disturbance drivers of plant community structure in the sagebrush steppe.

Abiotic conditions, biotic factors, and disturbances can act as filters that control community structure and composition. Understanding the relative importance of these drivers would allow us to understand and predict the causes and consequences of changes in community structure. We used long-term data (1989-2002) from the sagebrush steppe in the state of Washington, USA, to ask three questions: (1) What are the key drivers of community-level metrics of community structure? (2) Do community-level metrics and functional groups differ in magnitude or direction of response to drivers of community structure? (3) What is the relative importance of drivers of community structure? The vegetation in 2002 was expressed as seven response variables: three community-level metrics (species richness, total cover, compositional change from 1989 to 2002) and the relative abundances of four functional groups. We used a multi-model inference framework to identify a set of top models for each response metric beginning from a global model that included two abiotic drivers, six disturbances, a biotic driver (initial plant community), and interactions between the disturbance and biotic drivers. We also used a permutational relative variable importance metric to rank the influence of drivers. Moisture availability was the most important driver of species richness and of native forb cover. Fire was the most important driver of shrub cover and training area usage was important for compositional change, but disturbances, including grazing, were of secondary importance for most other variables. Biotic drivers, as represented by the initial plant communities, were the most important driver for total cover and for the relative covers of exotics and native grasses. Our results indicate that the relative importance of drivers is dependent on the choice of metric, and that drivers such as disturbance and initial plant community can interact.

English

The mined area in the Brazilian Federal District - BFD is proportionately five times larger than the national average, and the restoration of plant communities at these sites is both necessary and mandatory. The use woody species to restore degraded ecosystems is a common practice, and this study aimed to identify and assess the floristic composition of woody species introduced in mining sites of the BFD as a mean of ecosystem restoration. The floristic survey was conducted by sampling of groups of one hundred individuals, repeated until sampling sufficiency was achieved. 4,500 plants were sampled in ten sites, which housed 92 autochthonous and 21 allochthonous woody species. The plant communities surveyed consisted of between 13 to 62 species, planted at low density - 467 ± 222 plants ha-1. The preferential use of autochthonous species in the revegetation projects resemble the standards recommended by successional restoration models, but the floristic diversity and plant density in these sites fall below the values deemed ideal. The 92 autochthonous species include representatives of various habitats, ecological groups and dispersal syndromes. These 92 species currently in use may be regrouped in initial plant communities denser and more diverse than those found in the revegetated sites. The plant communities in the surveyed sites were composed almost exclusively of arboreal species and such pattern may lead succession towards the formation of ecosystems dissimilar to the original savanna formations present in the studied locations.

Erosion control blankets, organic amendments and site variability influenced the initial plant community at a limestone quarry in the Canadian Rocky Mountains

Abstract. Season of seeding and soil amendment with manure mix, wood shavings and erosion control blankets were evaluated over two growing seasons to determine their effect on soil properties and native grass establishment at a Canadian limestone quarry and lime processing plant. A season (fall, spring) of soil amending and seeding did not significantly affect revegetation or soil properties. Site characteristics such as slope, aspect, initial soil nutrients and surrounding plant communities influenced early plant community development and overall effects of soil treatments. Erosion control blankets resulted in the highest seeded plant cover and the lowest non-seeded plant cover despite not significantly changing soil chemical properties. Total nitrogen and carbon significantly increased establishment of seeded grasses and non-seeded species. Increased nitrogen and carbon in the constructed soils were best achieved through addition of manure. Wood shavings did not favour establishment of vegetation and resulted in similar, and in some cases less, vegetation than the controls. Assisted revegetation increased plant cover from &lt; 6 to 50% and reduced cover of non-seeded species. Amendments that modified both chemical and physical soil conditions were best to increase vegetation establishment in the harsh conditions of the quarry.

Effects of Warming on Shrub Abundance and Chemistry Drive Ecosystem-Level Changes in a Forest–Tundra Ecotone

Tundra vegetation is responding rapidly to on-going climate warming. The changes in plant abundance and chemistry might have cascading effects on tundra food webs, but an integrated understanding of how the responses vary between habitats and across environmental gradients is lacking. We assessed responses in plant abundance and plant chemistry to warmer climate, both at species and community levels, in two different habitats. We used a long-term and multisite warming (OTC) experiment in the Scandinavian forest–tundra ecotone to investigate (i) changes in plant community composition and (ii) responses in foliar nitrogen, phosphorus, and carbon-based secondary compound concentrations in two dominant evergreen dwarf-shrubs (Empetrum hermaphroditum and Vaccinium vitis-idaea) and two deciduous shrubs (Vaccinium myrtillus and Betula nana). We found that initial plant community composition, and the functional traits of these plants, will determine the responsiveness of the community composition, and thus community traits, to experimental warming. Although changes in plant chemistry within species were minor, alterations in plant community composition drive changes in community-level nutrient concentrations. In view of projected climate change, our results suggest that plant abundance will increase in the future, but nutrient concentrations in the tundra field layer vegetation will decrease. These effects are large enough to have knock-on consequences for major ecosystem processes like herbivory and nutrient cycling. The reduced food quality could lead to weaker trophic cascades and weaker top down control of plant community biomass and composition in the future. However, the opposite effects in forest indicate that these changes might be obscured by advancing treeline forests.

Vegetation trends following fire in the Roggeveld, Mountain Renosterveld, South Africa

The Mountain Renosterveld vegetation of the Roggeveld is an escarpment type renosterveld showing strong karroid affinities. Fire plays an important role as a landscape scale disturbance that shapes plant communities in this vegetation type, however, post-fire succession has never before been documented for renosterveld vegetation. A study was therefore conducted in the northern Roggeveld to improve our understanding of the recovery of the vegetation following fire. The natural vegetation recovery was analysed using line transect data accumulated at five different sites over a ten year period. This paper reports on the post-fire vegetation trends with respect to changes in species composition, species richness, life form composition and life form richness. Vegetation cover began to re-establish within the first nine months following the fire, and remained at a high level from years 3 to 10. At the first survey the species richness varied from 13 to 17 species, with the highest species richness (14 to 31 species) generally encountered at each transect after three years. The highest Shannon index values were generally found within the first three years and the lowest Shannon index values were found in years 9 and 10. In all cases the Principal Co-ordinate Analysis ordinations of the species composition data indicated a clear separation in the species composition between the first two years (years 1 and 2) following the fire and the remaining years (year 3 to 10). This study also supports the ‘initial floristic composition’ model of Egler (1954) in that all or the majority of species encountered during the succession were already present at the beginning of the recovery phase and there was a rapid re-establishment of the initial plant community.

A Method for Evaluating Outcomes of Restoration When No Reference Sites Exist

AbstractEcological restoration typically seeks to shift species composition toward that of existing reference sites. Yet, comparing the assemblages in restored and reference habitats assumes that similarity to the reference habitat is the optimal outcome of restoration and does not provide a perspective on regionally rare off‐site species. When no such reference assemblages of species exist, an accurate assessment of the habitat affinities of species is crucial. We present a method for using a species by habitat data matrix generated by biodiversity surveys to evaluate community responses to habitat restoration treatments. Habitats within the region are rated on their community similarity to a hypothetical restored habitat, other habitats of conservation concern, and disturbed habitats. Similarity scores are reinserted into the species by habitat matrix to produce indicator (I) scores for each species in relation to these habitats. We apply this procedure to an open woodland restoration project in north Mississippi (U.S.A.) by evaluating initial plant community responses to restoration. Results showed a substantial increase in open woodland indicators, a modest decrease in generalists historically restricted to floodplain forests, and no significant change in disturbance indicators as a group. These responses can be interpreted as a desirable outcome, regardless of whether species composition approaches that of reference sites. The broader value of this approach is that it provides a flexible and objective means of predicting and evaluating the outcome of restoration projects involving any group of species in any region, provided there is a biodiversity database that includes habitat and location information.

Impact of deer browsing on plant communities in cutover sites on Anticosti Island

Browsing by Anticosti Island's large white-tailed deer (Odocoileus virginianus) population reduces the abundance of the deer's preferred species and increases the abundance of less preferred species. This alteration of plant communities can lead to degradation of deer habitat. However, the centres of large clear-cuts may be spared from deer browsing given that this animal's instincts regarding predation risk impel it to remain near escape cover. This project assessed the impact of high deer populations on plant communities in large clear-cuts harvested by cutting with protection of regeneration and soils (CPRS). To this end, vegetation growing within fenced-off areas was compared with that growing in adjacent unfenced plots following application of large CPRS in 1995 and 1996. The results show that deer browsing does not have an impact on the number of plant species but does dramatically reduce total species cover. Furthermore, although the number of species was the same in the fenced and unfenced plots, some species were only found in one or the other type of plot. Rubus idaeus, Betula papyrifera, Epilobium angustifolium, Abies balsamea, Maianthemum canadense, Cornus Canadensis, and Prunus pensylvanica were all associated with fenced plots, while the cover of true grasses and Cirsium arvense increased in the unfenced plots. We conclude that large CPRS-harvested clear-cuts cannot serve to preserve the initial plant communities on Anticosti Island.

Relative contributions of adjacent species set and environmental conditions to plant community richness and composition in a group of created wetlands

A major topic of discussion in community ecology is the relative influence of proximate seed sources and environmental variability on local plant species diversity. We investigated the effect of adjacent seed sources and environmental factors on initial plant species richness and composition in sixty wetland basins created in central Kansas, USA in 1998. We defined the adjacent seed source for each basin as the list of plant species found within a 10-m radius around each basin, filtered for those species capable of growing in wetland conditions. Basin water levels were monitored biweekly over three growing seasons, starting in 1999, as were soil moisture, temperature, and conductivity. Soil samples were analyzed for total nitrogen, total phosphorus, pH, and soluble salts in 2001. We found that in the three-year-old basins, the species richness of the adjacent areas had contributed significantly to basin species richness, as did basin area, soil pH, and water-level fluctuation. After three years, 61% of the species found in the adjacent areas were also found in the basins. While we found that proximate seed sources did influence initial species richness and composition in newly created wetlands, we cannot conclude that any differences in initial plant communities will ultimately result in different successional trajectories for the wetlands in our system. However, given the potential sensitivity of vegetation richness and composition in these newly created wetlands to proximate seed sources, the location of future wetland creations may need to be considered.

PLANT COMMUNITY MEDIATED VS. NUTRITIONAL CONTROLS ON LITTER DECOMPOSITION RATES IN GRASSLANDS

Grasslands are one of the major biomes on earth and can serve as important soil carbon sinks. Nutrient enrichment of these grasslands can have a significant impact on carbon losses through the decomposition process. We investigated the effects of long-term (12-yr) experimentally increased N and/or P supply on litter production and on the chemistry and decomposition of bulk litter from two grasslands differing in soil nutrient status. Potential aboveground litter production in the controls of a P-rich low-productivity riparian grassland was lower than that in a N-rich high-productivity peat grassland and increased with enhanced N supply. Nutrient treatments did not enhance litter production in the high-productivity peat grassland. The concentrations of phenolics in bulk litter from the peat grassland, dominated by sedges, were 2-3 times higher than those in the riparian grassland that was dominated by herbs and grasses. At both sites increased nutrient supply had no detectable effect on phenolics concentrations. All P-related litter chemistry param- eters reflected the higher soil P status of the riparian grassland. P fertilization had a greater effect on litter chemistry in the P-deficient peat grassland than in the P-rich riparian grass- land. At both sites, there was no change in overall litter chemistry in response to N fer- tilization, except for higher lignin concentrations in the peat grassland litter. Both short- term (8 wk) litter incubations in the laboratory and a 3-yr litter bag study in the field showed that riparian grassland litter decomposed faster than litter from the peat grassland. The long-term nutrient additions had no significant effects on the decomposition of the bulk litter of each grassland type. Regression analysis on the combined data of the two sites showed that phenolics, and to a lesser extent, P-related litter chemistry parameters, exerted a strong control on both litter respiration and litter mass loss. Our study shows that long-term experimental nutrient additions do not lead to increased decomposition rates in grasslands and that the initial plant community litter quality is the main determinant of carbon losses through the decomposition process. The results of this study suggest that nutrient enrichment will likely affect ecosystem carbon balance more by affecting litter production than by affecting litter decomposition rates.

Potential contribution of the seed bank in coniferous plantations to the restoration of native deciduous forest vegetation

In this study, we compared the soil seed bank and current vegetation under coniferous plantations and adjacent native deciduous forests. The objective was to assess how much of the initial plant diversity is retained in such plantations, and the potential to restore this initial plant community from seed bank in case of reversion to broadleave stands. Four stands growing side by side and with different dominant species were selected at two locations (site of Haye: Quercus petraea, Pseudotsuga menziesii, Pinus sylvestris and Picea abies; site of La Petite-Pierre: Quercus petraea, Fagus sylvatica, Pinus sylvestris and Picea abies). In each stand, ground vegetation was surveyed and soil seed bank was sampled. Composition of ground flora and seed bank of stands were quite different: only 11 to 30 % of the species were in both the ground flora and the seed bank. Composition of the seed bank was mainly influenced by site location and sylvicultural practices such as the type of afforestation or the tree cover. Species richness of seed banks and vegetation were higher in the site of Haye than in the site of La Petite-Pierre. Seedling density strongly decreased with stand age. Whereas between 65 and 86 % of species found in the ground vegetation of native deciduous stand were also present in the understory or the seed bank of mature coniferous stands, this was only about 50 % in young coniferous stands. Species of deciduous stands which were absent from coniferous stands were typical of old forests. In contrast, species mainly found in the coniferous stands were often ruderal. In the studied areas, it would be possible to restore up to 86 % of the native deciduous forest vegetation, but some plant species typical of ancient forests may have disappeared during the coniferous stage.