Influence of four soil maintenance practices on Collembola communities in a Mediterranean vineyard

We studied the influence of four soil maintenance practices on Collembola communities in the soil of a Mediterranean vineyard: (a) postemergence herbicide with glyphosate; (b) postemergence and pre-emergence herbicides with glyphosate, terbuthylazine, diuron and oryzalin; (c) natural flora and (d) tillage to a depth of 10–15 cm. Total Collembola abundance, species diversity and species richness significantly varied between the four practices. Notably, the practice using postemergence and pre-emergence herbicides had significantly lower values. Identification of Collembola at species level allowed an interspecies comparison and revealed significant differences for the most common species between the four practices, with each practice being characterized by a different set of species. None of the species were found to be significantly more abundant in the plots treated with postemergence and pre-emergence herbicides.     

Soil microarthropods are only weakly impacted after 13 years of repeated drought treatment in wet and dry heathland soils

Studies of biological responses in the terrestrial environment to rapid changes in climate have mostly been concerned with aboveground biota, whereas less is known of belowground organisms. The present study focuses on mites and springtails of heathland ecosystems and how the microarthropod community has responded to simulated climate change in a long-term field experiment. Increased temperature and repeated drought was applied for 13 years to field plots located in Wales, The Netherlands and Denmark representing sites of contrasting climatic conditions with respect to precipitation and temperature. This approach provided an opportunity to study biological responses on a local (within sites) and regional scale. Warming treatments increasing night time temperature (0.3–1 °C higher than ambient at 5 cm soil depth) had no detectable effects on the microarthropod communities. Increased intensity and frequency of drought had only weak persistent effects on springtail species composition, but practically no effect on major mite groups (Oribatida, Prostigmata or Mesostigmata) suggesting that ecosystem functions of microarthropods may only be transiently impacted by repeated spring or summer drought.     

Food choice and reproductive success of Folsomia candida feeding on copper-contaminated mycelium of the soil fungus Alternaria alternata

We examined collembolan food preference for fungal mycelium grown on copper-contaminated medium, and the relationship between copper content, food selectivity and collembolan fitness when fed contaminated mycelium.To clarify whether collembolan food selectivity is related to fitness parameters, Folsomia candida were fed mycelium of the dark-pigmented fungus Alternaria alternata grown on medium with different copper concentrations. Copper-contaminated food (fungus grown on 50, 125, 250 and 500 μg Cu g⁻¹ medium, fresh wt.) was offered together with untreated food for 4 weeks. F. candida fed selectively on the provided mycelium and discriminated clearly between mycelium grown on high and low levels of contamination, distinctly preferring fungus grown on medium with a total copper concentration of 50 and 125 μg g⁻¹. In contrast, fungus grown on highly contaminated medium (250 and 500 μg g⁻¹) was avoided. Collembolan food preference generally matched fitness parameters. Reproduction was significantly affected by the total copper concentration of the fungal growth medium. When fed their preferred mycelium, collembolan reproduction was enhanced, whereas a diet of highly contaminated mycelium (250 or 500 μg g⁻¹) resulted in a strong decrease in reproduction. Adult survival was affected only marginally. Even though heavy metal contamination is a potential stress factor for many soil microarthropods, F. candida is able to discriminate between high and low quality food sources, and even benefits from moderately elevated copper concentrations.     

Chemical communication in Ceratophysella sigillata (Collembola: Hypogastruridae): intraspecific reaction to alarm substances

When disturbed or damaged, the winter-active collembolan Ceratophysella sigillata releases a strong aromatic odour, which can be perceived even by humans. The efficiency of the substances released as an intraspecific alarm substance was investigated. C. sigillata showed distinct reactions to chemicals released by conspecifics; crushed conspecifics, cuticle washings (with pentane) and whole body extracts (with methyl acetate) were tested. Our results demonstrate that a very sensitive intraspecific avoidance reaction is present to the substances released by crushed C. sigillata as well as to washings and extracts: extracts from as little as half a springtail induced a significant avoidance reaction in conspecifics. Alarm substances were present in all instars of C. sigillata including eggs, and all instars of C. sigillata tested showed similar reactions to the extracts or crushed conspecifics.     

Long-term dynamics of Collembola in two contrasting ecosystems

The dynamics of collembolan communities were investigated in a spruce forest and in a dry natural meadow (Moscow Region, Russia) from 1991 to 1998. The long-term dynamics of springtail communities were compared to understand the role of external and internal factors responsible for changes over time. The two springtail communities were similiar in species richness (54 and 52 species), species diversity, as well as their rank/dominance distribution patterns. There were two dominant species in the forest, Isotomiella minor (Schäff.) and Parisotoma notabilis (Schäff.), and three in the meadow, P. notabilis, Lepidocyrtus lignorum (Fabr.) and Protaphorura armata Gisin. Community organization was much more stable over time in the forest than in the meadow. Population densities of the two dominant species changed synchronously and correlated with mean monthly temperature in the forest. In the meadow, densities of only some dominant species (P. notabilis and L. lignorum) correlated with each other and with temperature. It is supposed that a constant organization of collembolan communities through time arises as a result of successional changes over long periods with relatively stable conditions allowing the gradual formation of complexes of dominant species.     

Does forest type and vegetation patchiness influence horizontal distribution of soil Collembola in two neighbouring forest sites?

The influence of forest type and heterogeneity of understory vegetation on the horizontal distribution of soil living Collembola was studied in two neighbouring mountain forest sites—a 180-year old beech forest and a 70-year old natural spruce forest. Four homogenous patches with different understory vegetation were chosen within each forest site and sampled 12 times between 1997 and 1998. A total of 56 collembolan species were identified, 51 in the beech forest and 48 in the spruce site. Twenty-three species were rare—they were recorded with low constancy and density. Although both forest sites differed in soil type, humus form and soil chemical parameters, the species composition of their collembolan communities was quite similar (77% shared species). Nevertheless, soil collembolan communities of both forest sites were clearly delimited on both qualitative (presence–absence) and quantitative (in density of individual species) levels as well as in terms of total collembolan density. Mean collembolan density reached 26 650–44 030 ind/m² in the beech patches compared to 44 470–68 050 ind/m² found in the spruce patches. Considerably higher densities of several species in one forest site may indicate more suitable habitat. In spite of similar species composition and minor differences in qualitative parameters among different vegetation patches within one forest site, there was clear variation in density of individual species, particularly between patches with and without herb vegetation. This could reflect different microclimatic conditions, additional litter supply from herbs or indirect interactions of Collembola with their roots.     

Collembolan grazing affects the growth strategy of the cord-forming fungus Hypholoma fasciculare

Mycelia of cord-forming fungi show remarkable patterns of reallocation of biomass and nutrients indicating an important role of these, often extensive, organisms in the spatial translocation of energy and nutrients in forest soils. Despite the rich tradition of interaction studies between soil microarthropods and fungi, the spatial implications of these interactions, due to the potential growth responses of the fungi and to the translocation of energy and nutrients within the mycelial network, have been largely ignored. In this paper we analyse fungal growth responses in two-dimensional model systems composed of compressed soil, the cord-forming fungus Hypholoma fasciculare and three fungivorous Collembolan species. We hypothesised that (i) the highly co-ordinated nature of cord-forming fungi would lead to growth responses following collembolan grazing, and that, (ii) such changes are dependent on grazing intensity, and (iii) changes are dependent on the species grazing. Mycelial extent and hyphal cover decreased with increasing grazing density; at highest grazing density also the fractal dimension of the mycelial border decreased, indicating a less branched foraging front due to the regression of fine hyphae and the development of mycelial cords. Effects differed greatly between collembolan species although they exerted comparable grazing pressure (the smaller species were added in larger numbers according to their allometric size-metabolic rate relationships): while grazing by Folsomia candida resulted in less mycelial extension and hyphal cover, these variables were not affected when Proisotoma minuta and Hypogastrura cf. tullbergi grazed. The effects of a species mix suggested an additive effect of the component species. This shows that fungal mycelia may suffer from damage caused by few but large collembolans, affecting extension as well as coverage of the mycelium, but that fungi may compensate for the biomass loss caused by more but slightly smaller collembolans. In about 20% of the model systems H. fasciculare switched from a growth pattern with a broad contiguous foraging front and uniform growth in all directions to a pattern with fast growing sectors while other sectors stopped growth completely. The switch occurred in grazed systems exclusively; thus we interpret this observation as a fugitive response and as a strategy for quickly escaping from places where grazing pressure is experienced.     

Carbon fluxes in soil food webs of increasing complexity revealed by C labelling and C natural abundance

Soil food webs are mainly based on three primary carbon (C) sources: root exudates, litter, and recalcitrant soil organic matter (SOM). These C sources vary in their availability and accessibility to soil organisms, which could lead to different pathways in soil food webs. The presence of three C isotopes (¹²C, ¹³C and ¹⁴C) offers an unique opportunity to investigate all three C sources simultaneously. In a microcosm experiment we studied the effect of food web complexity on the utilization of the three carbon sources. We choose an incomplete three factorial design with (i) living plants, (ii) litter and (iii) food web complexity. The most complex food web consisted of autochthonous microorganisms, nematodes, collembola, predatory mites, endogeic and anecic earthworms. We traced C from all three sources in soil, in CO₂ efflux and in individual organism groups by using maize grown on soil developed under C₃ vegetation and application of ¹⁴C labelled ryegrass shoots as a litter layer. The presence of living plants had a much greater effect on C pathways than food web complexity. Litter decomposition, measured as ¹⁴CO₂ efflux, was decreased in the presence of living plants from 71% to 33%. However, living plants increased the incorporation of litter C into microbial biomass and arrested carbon in the litter layer and in the upper soil layer. The only significant effect of food web complexity was on the litter C distribution in the soil layers. In treatments with fungivorous microarthropods (Collembola) the incorporation of litter carbon into mineral soil was reduced. Root exudates as C source were passed through rhizosphere microorganisms to the predator level (at least to the third trophic level). We conclude that living plants strongly affected C flows, directly by being a source of additional C, and indirectly by modifying the existing C flows within the food web including CO₂ efflux from the soil and litter decomposition.