Effects of European Earthworm Invasion on Soil Characteristics in Northern Hardwood Forests of Minnesota, USA

European earthworms are colonizing worm-free hardwood forests across North America. Leading edges of earthworm invasion in forests of northern Minnesota provide a rare opportunity to document changes in soil characteristics as earthworm invasions are occurring. Across leading edges of earthworm invasion in four northern hardwood stands, increasing total earthworm biomass was associated with rapid disappearance of the O horizon. Concurrently, the thickness, bulk density and total soil organic matter content of the A horizon increased, and it’s percent organic matter and fine root density decreased. Different earthworm species assemblages influenced the magnitude and type of change in these soil parameters. Soil N and P availability were lower in plots with high earthworm biomass compared to plots with low worm biomass. Decreases in soil nitrogen availability associated with high earthworm biomass were reflected in decreased foliar nitrogen content for Carex pensylvanica, Acer saccharum and Asarum canadense but increased foliar N for Athyrium felix-femina. Overall, high earthworm biomass resulted in increased foliar carbon to nitrogen ratios. The effects of earthworm species assemblages on forest soil properties are related to their feeding and burrowing habits in addition to effects related to total biomass. The potential for large ecosystem consequences following exotic earthworm invasion has only recently been recognized by forest ecologists. In the face of rapid change and multiple pressures on native forest ecosystems, the impacts of earthworm invasion on forest soil structure and function must be considered.     

Spatial and Temporal Dynamics of Exotic Earthworm Communities Along Invasion Fronts in a Temperate Hardwood Forest in South-Central New York (USA)

The invasion of North American forests by exotic earthworms is producing profound ecosystem changes, such as alterations in soil nutrient cycling, and redistribution and loss of soil organic matter. However, the present and future extent of these invasions is difficult to evaluate without a better understanding of the factors that control the distribution and abundance of earthworms in previously non-invaded habitats. In this study, the species composition and short-term dynamics of three exotic earthworm invasion fronts were studied at a northern hardwood forest in south-central New York State (USA). Belt transects were established at each of the three locations to sample from earthworm-invaded areas through transition zones and into invasion front areas. Lumbricus rubellus, L. terrestrisandOctolasion tyrtaeum were the most common species, but their distribution was not homogeneous along the transects. Whereas, L. rubellus was the only species with relatively high adult densities at transition zones and invasion fronts, L. terrestris and O. tyrtaeum occurred mostly in the heavily earthworm-invaded areas and were rare at the invasion fronts. The density of earthworms along the transects decreased by 60–87 from June 2001 to October 2002 and then recovered in 2003 to values similar to those of 2001. This decrease was apparently caused by reduced recruitment of immature earthworms, probably related to the severe drought periods that the study area experienced in 2001 and 2002. Our data suggest that climate and topography, through their effects on soil moisture patterns, can be critical factors controlling the distribution and spread of exotic earthworms in previously non-invaded habitats.     

Exotic Ecosystem Engineers Change the Emergence of Plants from the Seed Bank of a Deciduous Forest

The anthropogenic spread of exotic ecosystem engineers profoundly impacts native ecosystems. Exotic earthworms were shown to alter plant community composition of the understory of deciduous forests previously devoid of earthworms. We investigated the effect of two exotic earthworm species (Lumbricus terrestris L. and Octolasion tyrtaeum Savigny) belonging to different ecological groups (anecic and endogeic) on the emergence of plants from the seed bank of a northern North American deciduous forest using the seedling emergence method. We hypothesized that (1) exotic earthworms change the seedling emergence from the plant seed bank, (2) L. terrestris increases the emergence of plant seedlings of the deeper soil layer but decreases that of the upper soil layer due to plant seed burial, and (3) O. tyrtaeum decreases plant seedling emergence due the damage of plant seeds. Indeed, exotic earthworms altered the emergence of plant seedlings from the seed bank and the functional composition of the established plant seedlings. Surprisingly, although L. terrestris only marginally affected seedling emergence, O. tyrtaeum changed the emergence of native plant species from the seed bank considerably. In particular, the number of emerging grass and herb seedlings were increased in the presence of O. tyrtaeum in both soil layers. Moreover, the impacts of earthworms depended on the identity of plant functional groups; herb species benefited, whereas legumes suffered from the presence of exotic earthworms. The results highlight the strong effect of invasive belowground ecosystem engineers on aboveground ecosystem characteristics and suggest fundamental changes of ecosystems by human-spread earthworm species.     

Concordance of N bio-indicators: Dandelion and earthworms in a rotational pasture

Earthworms are important soil metabionts indicative of N enrichment in pastures. A rotational pasture in central Nova Scotia was tested for earthworms using chemical extraction followed by excavation and hand sorting in 28 paired micro plots placed in areas with low versus high proportion of the N indicator plant species dandelion (Taraxacum officinale). Species richness was low with five earthworm species of the Lumbricidae recovered in the following order of abundance: Lumbricus rubellus, Lumbricus terrestris, Aporrectodea turgida, Aporrectodea tuberculata, and Aporrectodea trapezoides. All species occurred at high constancy except the rare A. trapezoides. The inventory revealed spatial differentiation of earthworm abundance and community structure at the field level. High proportion of dandelion reduced pasture sward biomass while abundance of L. rubellus and A. tuberculata significantly (p < 0.05) increased with a concomitant increase in epigeic earthworm dominance at the expense of the anecic L. terrestris. Thus, low cost and non-destructive floristic surveys of N indicators, such as dandelion, allow for concordant inferences about the environmental impact of intensive cow pasture on earthworms and ecosystem function. High earthworm counts may run contrary to the notion of ecological integrity depending on specific earthworm abundances. Reduced earthworm benefits due to any de-intensification of rotational pasture must be assessed against increased risks of N-leaching in intensive pastures with high proportion of dandelion.     

Microsatellite markers for the earthworm Lumbricus rubellus

We describe eight microsatellite markers for the earthworm Lumbricus rubellus, a commonly used ‘biological sentinel’ species in ecotoxicology studies. Thirty‐four individuals from a single population were tested for polymorphism. Markers possessed between two and 15 alleles per locus, and expected heterozygosity ranged between 0.06 and 0.92 (observed heterozygosity 0.03–0.92). Unfortunately, no consistent cross‐species amplification was observed in the related congeners Lumbricus castaneus and Lumbricus terrestris.     

The impact of the ‘New Zealand flatworm’, Arthurdendyus triangulatus, on earthworm populations in the field

The ‘New Zealand flatworm’, Arthurdendyus triangulatus, is a native of the South Island of New Zealand, which has established in the UK, Ireland and the Faroe Islands. In its introduced range, it is a predator of lumbricid earthworms. To assess the impact of A. triangulatus on earthworm species, flatworm distributions were manipulated into ‘high’, control and ‘low’ densities within a replicated field experiment. Earthworm biomass in the ‘high’ flatworm density treatment was significantly lower than the control or ‘low’ treatments. This was due to a reduction in the anecic species Lumbricus terrestris and, to a lesser extent, Aporrectodea longa. There was little evidence of negative effects on other earthworm species, with even a weakly positive relationship between flatworm density and epigeic biomass. Principal components analysis showed a clear separation of anecic species from A. triangulatus, but the epigeic species Lumbricus festivus and Lumbricus rubellus grouped with A. triangulatus, suggesting that they could be benefitting from reduced intraguild competition. Flatworm densities of 0.8 per m², comparable to natural infestations in grassland, were predicted to give a reduction in total earthworm biomass of c. 20 %. The bulk of this was comprised of a reduction in anecic species biomass. In particular, it is considered that A. triangulatus poses a serious risk to L. terrestris populations, with implications for soil functioning and indigenous earthworm-feeding wildlife.     

Tree leaf litter composition and nonnative earthworms influence plant invasion in experimental forest floor mesocosms

Dominant tree species influence community and ecosystem components through the quantity and quality of their litter. Effects of litter may be modified by activity of ecosystem engineers such as earthworms. We examined the interacting effects of forest litter type and earthworm presence on invasibility of plants into forest floor environments using a greenhouse mesocosm experiment. We crossed five litter treatments mimicking historic and predicted changes in dominant tree composition with a treatment of either the absence or presence of nonnative earthworms. We measured mass loss of each litter type and growth of a model nonnative plant species (Festuca arundinacea, fescue) sown into each mesocosm. Mass loss was greater for litter of tree species characterized by lower C:N ratios. Earthworms enhanced litter mass loss, but only for species with lower C:N, leading to a significant litter × earthworm interaction. Fescue biomass was significantly greater in treatments with litter of low C:N and greater mass loss, suggesting that rapid decomposition of forest litter may be more favorable to understory plant invasions. Earthworms were expected to enhance invasion by increasing mass loss and removing the physical barrier of litter. However, earthworms typically reduced invasion success but not under invasive tree litter where the presence of earthworms facilitated invasion success compared to other litter treatments where earthworms were present. We conclude that past and predicted future shifts in dominant tree species may influence forest understory invasibility. The presence of nonnative earthworms may either suppress of facilitate invasibility depending on the species of dominant overstory tree species and the litter layers they produce.     

Earthworm invasion into previously earthworm-free temperate and boreal forests

Earthworms are keystone detritivores that can influence primary producers by changing seedbed conditions, soil characteristics, flow of water, nutrients and carbon, and plant–herbivore interactions. The invasion of European earthworms into previously earthworm-free temperate and boreal forests of North America dominated by Acer, Quercus, Betula, Pinus and Populus has provided ample opportunity to observe how earthworms engineer ecosystems. Impacts vary with soil parent material, land use history, and assemblage of invading earthworm species. Earthworms reduce the thickness of organic layers, increase the bulk density of soils and incorporate litter and humus materials into deeper horizons of the soil profile, thereby affecting the whole soil food web and the above ground plant community. Mixing of organic and mineral materials turns mor into mull humus which significantly changes the distribution and community composition of the soil microflora and seedbed conditions for vascular plants. In some forests earthworm invasion leads to reduced availability and increased leaching of N and P in soil horizons where most fine roots are concentrated. Earthworms can contribute to a forest decline syndrome, and forest herbs in the genera Aralia, Botrychium, Osmorhiza, Trillium, Uvularia, and Viola are reduced in abundance during earthworm invasion. The degree of plant recovery after invasion varies greatly among sites and depends on complex interactions with soil processes and herbivores. These changes are likely to alter competitive relationships among plant species, possibly facilitating invasion of exotic plant species such as Rhamnus cathartica into North American forests, leading to as yet unknown changes in successional trajectory.     

Root Foraging Influences Plant Growth Responses to Earthworm Foraging

Interactions among the foraging behaviours of co-occurring animal species can impact population and community dynamics; the consequences of interactions between plant and animal foraging behaviours have received less attention. In North American forests, invasions by European earthworms have led to substantial changes in plant community composition. Changes in leaf litter have been identified as a critical indirect mechanism driving earthworm impacts on plants. However, there has been limited examination of the direct effects of earthworm burrowing on plant growth. Here we show a novel second pathway exists, whereby earthworms (Lumbricus terrestris L.) impact plant root foraging. In a mini-rhizotron experiment, roots occurred more frequently in burrows and soil cracks than in the soil matrix. The roots of Achillea millefolium L. preferentially occupied earthworm burrows, where nutrient availability was presumably higher than in cracks due to earthworm excreta. In contrast, the roots of Campanula rotundifolia L. were less likely to occur in burrows. This shift in root behaviour was associated with a 30% decline in the overall biomass of C. rotundifolia when earthworms were present. Our results indicate earthworm impacts on plant foraging can occur indirectly via physical and chemical changes to the soil and directly via root consumption or abrasion and thus may be one factor influencing plant growth and community change following earthworm invasion. More generally, this work demonstrates the potential for interactions to occur between the foraging behaviours of plants and soil animals and emphasizes the importance of integrating behavioural understanding in foraging studies involving plants.     

Nonlinearity of effects of invasive ecosystem engineers on abiotic soil properties and soil biota

Invasions of non‐indigenous species into natural communities are currently rated as one of the most important threats to biodiversity. Particularly exotic ecosystem engineers such as earthworms potentially have profound impacts on community assembly and functioning. We investigated the impact of invasion by the lumbricid earthworms into an aspen forest of the Canadian Rocky Mountains on soil organic matter, microorganisms and microarthropod communities. Building on the results of previous studies in this forest, we expected positive effects of Lumbricus terrestris middens and negative effects of Octolasion tyrtaeum on soil biota (increase and decrease in soil nutrient concentrations, microbial parameters and soil microarthropod density and diversity, respectively). Further, we expected that earthworm effects change with time. Combined results of previous and the present study suggest a wavelike colonization pattern for Dendrobaena octaedra and O. tyrtaeum and that indeed the impact of earthworms on soil biota changed with time, likely due to changes in earthworm density. Unexpectedly, L. terrestris middens neither affected soil abiotic nor soil biotic properties. By contrast and in contrast to our hypothesis, carbon and nitrogen concentration and C‐to‐N ratio in deeper soil layers increased in presence of O. tyrtaeum, thereby likely enhancing nutrient availability for soil microorganisms and microarthropods. Even though the density of this endogeic species was rather low, presence of O. tyrtaeum resulted in increased densities of a number of microarthropod taxa and increased microarthropod diversity. The results suggest that at low density, invasive ecosystem engineers, such as O. tyrtaeum, cause disturbances of intermediate strength thereby beneficially affecting soil microorganisms and most microarthropods. This contrasts earlier effects during the wavelike invasion of O. tyrtaeum into the aspen forest when densities of O. tyrtaeum were high resulting in generally detrimental effects on soil biota. The results emphasize the nonlinearity of earthworm effects on abiotic and biotic soil properties and call for further long‐term investigations.     

Earthworm invasions of ecosystems devoid of earthworms: effects on soil microbes

Recent studies document North American earthworm invasions and their profound effects on the structure of the soil profile, which is the habitat for soil microorganisms (mainly fungi and bacteria). Dramatic alterations made to these layers during earthworm invasion significantly change microbial community structure and therefore microbial activities such as C transformations. Understanding the impacts of earthworm invasion on the microbes themselves will give insight into earthworm effects on microbial activities. Bacterial and actinomycete communities in earthworm guts and casts have not been studied in environments recently invaded by earthworms. Earthworm invasion tended to decrease fungal species density and fungal species diversity and richness. The presence of earthworms decreased zygomycete species abundance probably due to disruption of fungal hyphae. Physical disruption of hyphae may also explain decreased mycorrhizal colonization rates, decreased mycorrhizal abundance and altered mycorrhizal morphology in the presence of earthworms. Mixing of organic layers into mineral soil during earthworm invasion tended to decrease microbial biomass in forest floor materials while increasing it in mineral soil. In newly invaded forest soils, microbial respiration and the metabolic quotient tended to decline. In forests where either the microbial community has had time to adapt to earthworm activities, or where the destruction of the forest floor is complete, as in invasions by the Asian Amynthas hawayanus, the presence of earthworms tends to increase the metabolic quotient indicating a shift to a smaller, more active microbial community.     

Lumbricid earthworm invasion in the Carpathian Mountains and some other sites in Romania

Recent field research shows that the ubiquitous Lumbricus terrestris and Octolasion lacteum (the large parthenogenetic form) have invaded new habitats in Romania, replacing mostly endemic or rare species of the former earthworm communities which were recorded 20–30 years ago. Seven different cases of earthworm invasion are currently described. Invader species have larger ecological tolerances as compared with the endemic species.     

Earthworm community structure and diversity in experimental agricultural watersheds in Northeastern Ohio

Earthworms are known to have an important impact on soil fertility but much remains to be known about the factors that influence earthworm abundance and species diversity in agricultural soils and the impact of earthworm diversity on soil processes in those soils. We have studied factors that influence earthworm community structure and biodiversity in experimental agricultural watersheds at the North Appalachian Experimental Watershed near Coshocton, Ohio. We sampled earthworm communities in seven such watersheds from 1990 to 1992. Six earthworm species were present:Aporrectodea caliginosa, A. trapezoides, A. tuberculata, Lumbricus rubellus, Lumbricus terrestris and Octolasion tyrtaeum. The total earthworm biomass ranged from 2 to 32 g m^-2 and population levels ranged from 10 to 350 worms m^-2. Earthworm community structure and diversity differed among watersheds and was influenced by cropping patterns, geographic location and tillage. The greatest earthworm diversity and highest earthworm population levels occurred in a no-tillage watershed and a watershed that had previously been in ryegrass and long-term no-till. High earthworm populations were also observed in a chisel-plowed watershed. Watersheds that had been managed identically for 50 years had much different earthworm communities, indicating that factors other than management were important in determining species composition. The three watersheds with the lowest populations and diversity were adjacent to one another at one end of the station, indicating that at this site geographic location had the predominant influence on earthworm communities. A severe drought in 1991 greatly reduced earthworm populations and biomass. However, earthworm species differed in their response to drought withLumbricus rubellus appearing to be the most drought-sensitive andAporrectodea spp. the most drought-tolerant.     

Residue incorporation depth is a controlling factor of earthworm‐induced nitrous oxide emissions

Earthworms can increase nitrous oxide (N₂O) emissions, particularly in no‐tillage systems where earthworms are abundant. Here, we study the effect of residue incorporation depth on earthworm‐induced N₂O emissions. We hypothesized that cumulative N₂O emissions decrease with residue incorporation depth, because (i) increased water filled pore space (WFPS) in deeper soil layers leads to higher denitrification rates as well as more complete denitrification; and (ii) the longer upward diffusion path increases N₂O reduction to N₂. Two 84‐day laboratory mesocosm experiments were conducted. First, we manually incorporated maize (Zea mays L.) residue at different soil depths (incorporation experiment). Second, ¹³C‐enriched maize residue was applied to the soil surface and anecic species Lumbricus terrestris (L.) and epigeic species Lumbricus rubellus (Hoffmeister) were confined to different soil depths (earthworm experiment). Residue incorporation depth affected cumulative N₂O emissions in both experiments (P < 0.001). In the incorporation experiment, N₂O emissions decreased from 4.91 mg N₂O–N kg^?1 soil (surface application) to 2.71 mg N₂O–N kg^?1 soil (40–50 cm incorporation). In the earthworm experiment, N₂O emissions from L. terrestris decreased from 3.87 mg N₂O–N kg^?1 soil (confined to 0–10 cm) to 2.01 mg N₂O–N kg^?1 soil (confined to 0–30 cm). Both experimental setups resulted in dissimilar WFPS profiles that affected N₂O dynamics. We also found significant differences in residue C recovery in soil organic matter between L. terrestris (28–41%) and L. rubellus (56%). We conclude that (i) N₂O emissions decrease with residue incorporation depth, although this effect was complicated by dissimilar WFPS profiles; and (ii) larger residue C incorporation by L. rubellus than L. terrestris indicates that earthworm species differ in their C stabilization potential. Our findings underline the importance of studying earthworm diversity in the context of greenhouse gas emissions from agro‐ecosystems.     

Exotic earthworms dispersion through protected forest areas and their potential impacts on nitrous oxide production

It is generally accepted that human activities are responsible for the dispersal of exotic earthworms in northeastern North America. We know little, however, about the relative effects of concurrent human activities on the structure of these earthworm communities in protected forest areas, nor on their impacts on soil biological activities. Our first objective was to infer the relative importance of recreational fishing and road traffic on the structure of Lumbricidae communities in Mont-Tremblant National Park, the oldest conservation area in the province of Quebec, Canada. Our second objective was to test the relationship between earthworm species abundances and soil properties related to microbial and nitrogen dynamics. We sampled earthworm communities around 61 lakes, which included 23 heavily-fished lakes and 20 non-fished lakes located near roads, as well as 18 non-fished lakes located in remote areas of the park. Our results revealed that fishing and proximity to roads both have a positive effect on the abundance of earthworms, as does the soil pH. Fishing activities had a greater effect than road proximity on the abundance and diversity of earthworm communities, notably on the abundance of the anecic species Lumbricus terrestris. To assess at a finer scale the effects of earthworm community structure on soil microbial and nitrogen dynamics, we collected and analyzed soils from 47 sampling points around two lakes with high earthworm densities. Exploratory redundancy analysis found a negative correlation between epigeic and anecic earthworm species, with the former correlating positively to microbial biomass and the latter correlating positively to nitrification and denitrification. Confirmatory path analysis established a positive indirect effect of Lumbricus terrestris, the preferred fishing bait, on potential soil nitrous oxide emissions. We conclude that the human-mediated dispersion of earthworms in the most pristine ecosystems of Quebec affects ecosystem functioning and thus requires a review of current policies regarding the use of live-bait by fishermen.     

Earthworm effects on plant growth do not necessarily decrease with soil fertility

Earthworms are known to generally increase plant growth. However, because plant-earthworm interactions are potentially mediated by soil characteristics the response of plants to earthworms should depend on the soil type. In a greenhouse microcosm experiment, the responsiveness of plants (Veronica persica, Trifolium dubium and Poa annua) to two earthworm species (in combination or not) belonging to different functional groups (Aporrectodea. caliginosa an endogeic species, Lumbricus terrestris an anecic species) was measured in term of biomass accumulation. This responsiveness was compared in two soils (nutrient rich and nutrient poor) and two mineral fertilization treatments (with and without). The main significant effects on plant growth were due to the anecic earthworm species. L. terrestris increased the shoot biomass and the total biomass of T. dubium only in the rich soil. It increased also the total biomass of P. annua without mineral fertilization but had the opposite effect with fertilization. Mineral fertilization, in the presence of L. terrestris, also reduced the total biomass of V. persica. L. terrestris did not only affect plant growth. In P. annua and V. persica A. caliginosa and L. terrestris also affected the shoot/root ratio and this effect depended on soil type. Finally, few significant interactions were found between the anecic and the endogeic earthworms and these interactions did not depend on the soil type. A general idea would be that earthworms mostly increase plant growth through the enhancement of mineralization and that earthworm effects should decrease in nutrient-rich soils or with mineral fertilization. However, our results show that this view does not hold and that other mechanisms are influential.     

Leaf Litter Disappearance in Earthworm-Invaded Northern Hardwood Forests: Role of Tree Species and the Chemistry and Diversity of Litter

Earthworm invasion in North American temperate forest reduces forest floor mass, yet the interactions between litter composition, invasive earthworm community composition, and forest floor structure and composition are not well understood. For 2?years, we compared disappearance of leaf litter in field mesocosms in which we manipulated litter composition (monocultures of Quercus rubra, Acer saccharum, and Tilia americana litter, and an equal mixture of all three) and thereby the initial litter chemistry (C, C fractions, N, Ca) in sites with and without the major litter-feeding invasive earthworm species. The disappearance of litter mass followed the same ranking at both the sites: T. americana?>?equal mixtures?>?A. saccharum?≥?Q. rubra. However, differences in disappearance rate between the sites depended on litter composition and time. The differences in mass loss among litters of different compositions were greatest at the site invaded by the large litter-feeding earthworm, Lumbricus terrestris, and especially for T. americana and the mixture. Similarly, observed disappearance of the litter mixture was faster than predicted by an additive model at the site with L. terrestris, especially for the higher quality litter component in early summer. Initial litter calcium content was the best predictor (R ²?≥?0.90) of overall litter mass remaining each year, supporting the idea of the importance of calcium in forest floor dynamics, especially in the presence of calciferous, invasive earthworms.     

The post-registration monitoring of glyphosate-treated plants using anecic earthworms

Glyphosate N-(phosphonomethyl) glycine is a widely-used herbicide in agriculture. The anecic earthworm, Lumbricus terrestris feeds and forages for surface plant materials meaning that this species has a unique and direct exposure to agrichemicals. At the recommended product rates, significantly (F_1,44 = 8.67, p = .005) higher numbers of L. terrestris middens were found in the glyphosate treated areas of an arable crop field. Laboratory feeding assays using field aged plant materials indicated that previous glyphosate treatment was a statistically significant factor affecting earthworm L. terrestris biomass (F₁,₁₂ = 5.75, p = .03). Negligible glyphosate residues were detectable, and the field aged plant materials were encrusted with fungal hyphae. This suggests that glyphosate influences the colonisation of plant material by a litter-fungus complex which improves the food quality to earthworms. Concentrations of epoxiconazole, a fungicide, were detected in some plant materials and may influence overall food quality to earthworms. Glyphosate treatment on fresh volunteer plant leaves (unwanted crop seedlings) was not a statistically significant factor affecting earthworm L. terrestris biomass (F₁,₆ = 0.16, p = .92). These results indicate fungal communities influence feeding behaviours, and plant materials are a direct source of agrichemicals to anecic earthworms.     

Invasibility of experimental grassland communities: the role of earthworms, plant functional group identity and seed size

Invasions of natural communities by non‐indigenous species threaten native biodiversity and are currently rated as one of the most important global‐scale environmental problems. The mechanisms that make communities resistant to invasions and drive the establishment success of seedlings are essential both for management and for understanding community assembly and structure. Especially in grasslands, anecic earthworms are known to function as ecosystem engineers, however, their direct effects on plant community composition and on the invasibility of plant communities via plant seed burial, ingestion and digestion are poorly understood. In a greenhouse experiment we investigated the impact of Lumbricus terrestris, plant functional group identity and seed size of plant invader species and plant functional group of the established plant community on the number and biomass of plant invaders. We set up 120 microcosms comprising four plant community treatments, two earthworm treatments and three plant invader treatments containing three seed size classes. Earthworm performance was influenced by an interaction between plant functional group identity of the established plant community and that of invader species. The established plant community and invader seed size affected the number of invader plants significantly, while invader biomass was only affected by the established community. Since earthworm effects on the number and biomass of invader plants varied with seed size and plant functional group identity they probably play a key role in seedling establishment and plant community composition. Seeds and germinating seedlings in earthworm burrows may significantly contribute to earthworm nutrition, but this deserves further attention. Lumbricus terrestris likely behaves like a ‘farmer’ by collecting plant seeds which cannot directly be swallowed or digested. Presumably, these seeds are left in middens and become eatable after partial microbial decay. Increased earthworm numbers in more diverse plant communities likely contribute to the positive relationship between plant species diversity and resistance against invaders.     

Experimental Evaluation of Herbivory on Live Plant Seedlings by the Earthworm Lumbricus terrestris L. in the Presence and Absence of Soil Surface Litter

Background

Recent studies suggested that the earthworm Lumbricus terrestris might act as a seedling predator by ingesting emerging seedlings, and individuals were observed damaging fresh leaves of various plant species in the field. To evaluate the significance of herbivore behavior of L. terrestris for plant and earthworm performance we exposed 23- to 33-days-old seedlings of six plant species to earthworms in two microcosm experiments. Plants belonged to the three functional groups grasses, non-leguminous herbs, and legumes. Leaf damage, leaf mortality, the number of leaves as well as mortality and growth of seedlings were followed over a period of up to 26 days. In a subset of replicates 0.1 g of soil surface litter of each of the six plant species was provided and consumption was estimated regularly to determine potential feeding preferences of earthworms.

Results

There was no difference in seedling growth, the number of live seedlings and dead leaves between treatments with or without worms. Fresh leaves were damaged eight times during the experiment, most likely by L. terrestris, with two direct observations of earthworms tearing off leaf parts. Another nine leaves were partly pulled into earthworm burrows. Lumbricus terrestris preferred to consume legume litter over litter of the other plant functional groups. Earthworms that consumed litter lost less weight than individuals that were provided with soil and live plants only, indicating that live plants are not a suitable substitute for litter in earthworm nutrition.

Conclusion

Our results demonstrate that L. terrestris damages live plants; however, this behavior occurs only rarely. Pulling live plants into earthworm burrows might induce microbial decomposition of leaves to make them suitable for later consumption. Herbivory on plants beyond the initial seedling stage may only play a minor role in earthworm nutrition and has limited potential to influence plant growth.