共查询到20条相似文献，搜索用时 15 毫秒
1.
Humus forms in two secondary semi-evergreen tropical forests 总被引：5，自引：0，他引：5
The dynamics and function of humus forms in tropical forests are still poorly understood. Humus profiles in two secondary semi‐evergreen woodlands in Guadeloupe (French West Indies) were analysed micromorphologically. The humus forms are described under the canopy of five dominant tree species at two sites: under Pisonia subcordata and Bursera simaruba in a secondary forest on a Leptosol (Rendzina), and under Swietenia macrophylla, Tabebuia heterophylla and B. simaruba in a plantation on a calcareous Vertisol. In the secondary forest, two distinct humus forms were observed. A calcareous Amphimull, characterized by an OH horizon comprising the faecal pellets of millipedes, is formed under the canopy of P. subcordata, which produces a litter that is rich in nitrogen. A Dysmull with a thick root mat (OFRh horizon) develops under the canopy of B. simaruba, which produces a litter rich in lignin and phenol that is consumed slowly by the soil fauna. In the plantation on the Vertisol, the activity of the endoanecic earthworm Polypheretima elongata has led to the rapid disappearance of litter and the mixing of organic and mineral material. The humus form is a Eumull and is similar under all three tree species present. 相似文献
2.
3.
Gladys Loranger-Merciris Daniel Imbert France Bernhard-Reversat Patrick Lavelle Jean-François Ponge 《Biology and Fertility of Soils》2008,45(1):93-98
Despite the impact of soil millipedes on litter fragmentation in tropical forests, there have been few studies dealing with
factors determining their habitat preference in these ecosystems. In a natural secondary dry forest of Guadeloupe on Leptosol,
two complementary studies were carried out in order to test the hypothesis that litter N-content strongly influences millipede
distribution. Millipede abundance and species richness were described in the field under two tree species, Bursera simaruba and Pisonia subcordata, and were related to the chemical characteristics of their foliage. In addition, a laboratory experiment was done in order
to assess millipede feeding preferences regarding the chemical characteristics of leaves from various species. Millipede abundance
and species richness were significantly higher under P. subcordata than under B. simaruba, probably due to the higher N content of P. subcordata leaves. Moreover, millipedes fed preferentially on N-rich leaves. The present study confirms that there was a close correlation
between the preferred food, its chemical composition and the local distribution of millipede populations. 相似文献
4.
The aim of this field experiment was to quantify the contribution of soil fauna to plant litter decomposition in three forest sites differing in C/N ratio under natural conditions in Xishuangbanna, southwestern China. We conducted a survey of soil fauna communities, the forest floor litter and investigated mass loss of mixed tree species leaf litter for two years in a tropical secondary forest, an evergreen broad-leaf forest and a tropical rain forest. Exclusion treatments of different sized soil fauna from the leaf litter by using varying mesh size litter bags (2 mm and 0.15 mm) were also performed. Mass loss and C and N concentrations in litter bag leaf materials were determined at monthly intervals. We found that: (1) the three forests differed in floor litter biomass and nutrient contents but not in soil fauna richness and abundance; (2) litter mass loss and decomposition rate were slower when soil macrofauna and most of mesofauna were excluded; and (3) greatest soil fauna contribution to plant litter decomposition occurred in the rain forest, where leaf litter C/N ratio was also highest (41.5% contribution: 54.8 C/N ratio), in comparison to 8.69% in the broad-leaf forest and 19.52% in the secondary forest, both with low leaf litter C/N ratios (<32). Our results suggested that, soil fauna played a more pronounced role in the decomposition of mixed leaf litter in tropical rain forest, and significantly bigger effects from fauna were ascribed to the enhancement of N concentration and decrease of C concentration of the initially high C/N ratio litter in this forest site. 相似文献
5.
Forest ecosystems have been widely fragmented by human land use. Fragmentation induces significant microclimatic and biological differences at the forest edge relative to the forest interior. Increased exposure to solar radiation and wind at forest edges reduces soil moisture, which in turn affects leaf litter decomposition. We investigate the effect of forest fragmentation, soil moisture, soil macrofauna and litter quality on leaf litter decomposition to test the hypothesis that decomposition will be slower at a forest edge relative to the interior and that this effect is driven by lower soil moisture at the forest edge. Experimental plots were established at Wytham Woods, UK, and an experimental watering treatment was applied in plots at the forest edge and interior. Decomposition rate was measured using litter bags of two different mesh sizes, to include or exclude invertebrate macrofauna, and containing leaf litter of two tree species: easily decomposing ash (Fraxinus excelsior L.) and recalcitrant oak (Quercus robur L.). The decomposition rate was moisture-limited at both sites. However, the soil was moister and decomposition for both species was faster in the forest interior than at the edge. The presence of macrofauna accelerated the decomposition rate regardless of moisture conditions, and was particularly important in the decomposition of the recalcitrant oak. However, there was no effect of the watering treatment on macrofauna species richness and abundance. This study demonstrates the effect of forest fragmentation on an important ecosystem process, providing new insights into the interacting effects of moisture conditions, litter quality, forest edge and soil macrofauna. 相似文献
6.
Mascha Jacob Christian Platner Christoph Leuschner 《Soil biology & biochemistry》2009,41(10):2122-119
In the litter of six deciduous tree species (Fagus sylvatica, Tilia spp., Fraxinus excelsior, Carpinus betulus, Acer pseudoplatanus and Acer platanoides) and in stand-specific litter mixtures, we compared mass loss and nutrient release across diversity levels along a gradient of decreasing proportion of Fagus in stands with similar environmental and physical soil conditions. The litterbag studies ran over 22 months. The decomposition rate constants (k) of the temperate forest species ranged from k = 0.5 for Fagus to k = 1.5-2 for all other tree species. In Fagus, k was closely negatively correlated with the thickness of the litter layer and positively correlated with soil pH and isopod abundance. k was significantly higher in the mixed species stands (except for Carpinus and Fraxinus) and was positively correlated with earthworm abundance. Over the whole incubation time, nutrient amount and release rates of N, P, K, Ca and Mg as well as C-related ratios showed significant differences between tree species but no consistent differences among the diversity levels. Initial C-related nutrient ratios of the leaf litter and abundance of mesofauna and macrofauna were correlated with the length of time lag before nutrient release. We conclude that the mere number of tree species is not the main driver of nutrient release and decomposition processes, but that key groups of the decomposer fauna as well as the characteristic traits of the individual tree species are decisive. 相似文献
7.
Although there has been much recent interest in the effect of litter mixing on decomposition processes, much remains unknown about how litter mixing and diversity affects the abundance and diversity of decomposer organisms. We conducted a litter mixing experiment using litterbags in a New Zealand rainforest, in which treatments consisted of litter monocultures of each of 8 forest canopy and understory plant species, as well as mixtures of 2, 4 and 8 species. We found litter mixing to have little effect on net decomposition rates after either 279 or 658 days, and for each species decomposition rates in mixture treatments were the same as in monoculture. Litter species identity had important effects on litter microfauna, mesofauna and macrofauna, with different litter types promoting different subsets of the fauna. Litter mixing had few effects on densities of mesofauna and macrofauna, but did have some important effects on components of the microfauna, notably microbe-feeding and predatory nematodes. At day 279, litter mixing also consistently reduced the ratio of bacterial-feeding to microbe-feeding (bacterial-feeding+fungal-feeding) nematodes, pointing to mixing causing a significant switch from the bacterial-based to the fungal-based energy channel. Litter mixing sometimes influenced the community composition and diversity of nematodes and macrofauna, but effects of litter mixing on diversity were not necessarily positive, and were much weaker than effects of litter species identity on diversity. We conclude that litter mixing effects on the abundance and diversity of decomposer biota, when they occur, are likely to be of secondary and generally minor significance when compared to the effects of litter species identity and composition. 相似文献
8.
Desert ecosystems are characterized by sparse vegetation that affects both abiotic parameters and soil biota along the soil profile.This study was conducted in 2010–2011 in a loess plain in the northern Negev Desert highlands, Israel, to test two main hypotheses:1) the abundance and diversity of microarthropods would vary seasonally in the top 30-cm soil layer, but would be relatively stable at soil depths between 30 and 50 cm and 2) soil microarthropods would be more abundant in soils under shrubs with large litter accumulations than under shrubs with less litter or bare soil. Soil samples were collected each season from the 0–50 cm profile at10-cm intervals under the canopies of Hammada scoparia and Zygophyllum dumosum and from the bare interspaces between them.Soil moisture and soil organic carbon in the top 30-cm layers varied seasonally, but there was little variation in the soil layers deeper than 30 cm. Soil mites were most abundant in the top 30-cm soil layer in autumn and winter, with the highest number of families found in winter. There were no differences in soil microarthropod abundance attributable to the presence or absence of shrubs of either species. The microarthropod communities of the microhabitats studied consisted of Acari, Psocoptera, and Collembola. The Acari were mostly identified to the family level and were dominated by Oribatida(55%) and Prostigmata(41%) in all seasons and microhabitats, while the psocopterans were most abundant in summer. These results are opposite to those obtained in other studies in similar xeric environments. Moreover, our findings were not in line with our hypothesis that a better microhabitat played a major role in microarthropod community composition, diversity, and density. 相似文献
9.
Forest wildfires can dramatically affect soil communities and reduce abundance and diversity of soil fauna. The recovery of soil animals after a fire depends both on immigration from the unburnt forest and on local survival in less-burnt spots, but the relative importance of these mechanisms is poorly known. Therefore, these factors were studied with regard to soil macrofauna and soil mites seven years after a wildfire occurring in a pine forest area with shallow soil in 2001 in central Sweden. Three replicate transects, each consisting of four plots were studied. The plots were located in (i) the unburnt forest close to the fire edge; (ii) slightly burnt patches directly attached to the unburnt forest; (iii) slightly burnt patches surrounded by bare rock but connected to the unburnt forest edge by a corridor with mostly unburnt litter and vegetation; and (iv) island patches not connected with a corridor to the unburnt forest edge. The hypothesis was that that soil animals would particularly disperse from the unburnt forest to moderately burnt plots inside a burnt area via the network of less-burnt corridors. Poor dispersers would be especially few in the island patches lacking connection to the “mainland”, whereas good dispersers would be independent of gaps in connectivity. As expected, the highest abundance of both macrofauna and oribatid mites was found in the unburnt forest. Resident soil macro- and mesofauna representatives had half the abundance in the edge and corridor plots as compared to the control, but their abundance was not lower in the island plots than in the corridor plots indicating on-site survival and recovery. Mobile mesostigmatid mites did not show any significant reduction of abundance in any of the plots. The abundance of soil-dwelling oribatid mites did not differ between islands and unburnt forest, but mobile aboveground oribatids had significantly lower abundance on the islands than in the unburnt forest. The opposite was observed for aboveground and belowground oribatid mite species richness. In conclusion, belowground animals showed mainly local survival and seemed to be independent of corridors presence, whereas most aboveground and mobile macro- and mesofauna seemed to be more responsive to isolation induced by forest fires. Soil and litter corridors connecting unburnt patches inside the burnt forests with the unburnt edges were important mainly for less mobile groups of soil macrofauna. This supports the idea that there is a relatively slow process of soil ecosystem recovery and that external colonization of the burnt areas dominates over the local survival and recovery from refuges. 相似文献
10.
The rhizosphere and the surrounding soil harbor an enormous microbial diversity and a specific community structure, generated by the interaction between plant roots and soil bacteria. The aim of this study was to address the influences of tree species, tree species diversity and leaf litter on soil bacterial diversity and community composition. Therefore, mesocosm experiments using beech, ash, lime, maple and hornbeam were established in 2006, and sampled in October 2008 and June 2009. Mesocosms were planted with one, three or five different tree species and treated with or without litter overlay.Cluster analysis of DGGE-derived patterns revealed a clustering of 2008 sampled litter treatments in two separated clusters. The corresponding treatments sampled in 2009 showed separation in one cluster. PCA analysis based on the relative abundance of active proteobacterial classes and other phyla in beech and ash single-tree species mesocosm indicated an effect of sampling time and leaf litter on active bacterial community composition. The abundance of next-generation sequencing-derived sequences assigned to the Betaproteobacteria was higher in the litter treatments, indicating a higher activity, under these conditions. The Deltaproteobacteria, Nitrospira and Gemmatimonadetes showed an opposite trend and were more active in the mesocosms without litter. The abundance of alphaproteobacterial sequences was higher in mesocosms sampled in 2009 (P = 0.014), whereas the Acidobacteria were more active in 2008 (P = 0.014). At the family level, we found significant differences of the litter vs. non-litter treated group. Additionally, an impact of beech and ash as tree species on soil bacterial diversity was confirmed by the Shannon and Simpson indices. Our results suggest that leaf litter decomposition in pH-stable soils affect the soil bacterial composition, while tree species influence the soil bacterial diversity. 相似文献
11.
Trees interact in a complex manner with soils: they recycle and redistribute nutrients via many ecological pathways. Nutrient distribution via leaf litter is assumed to be of major importance. Beech is commonly known to have lower nutrient concentrations in its litter than other hardwood tree species occurring in Central Europe. We examined the influences of distribution of beech (Fagus sylvatica L.), ash (Fraxinus excelsior L.), lime (Tilia cordata Mill. and T. platyphyllos Scop.), maple (Acer spp. L.), and clay content on small‐scale variability of pH and exchangeable Ca and Mg stocks in the mineral soil and of organic‐C stocks in the forest floor in a near‐natural, mature mixed deciduous forest in Central Germany. The soil is a Luvisol developed in loess over limestone. We found a positive effect of the proportion of beech on the organic‐C stocks in the forest floor and a negative effect on soil pH and exchangeable Ca and Mg in the upper mineral soil (0 to 10 cm). The proportion of ash had a similar effect in the opposite direction, the other species did not show any such effect. The ecological impact of beech and ash on soil properties at a sample point was explained best by their respective proportion within a radius of 9 to 11 m. The proportion of the species based on tree volume within this radius was the best proxy to explain species effects. The clay content had a significant positive influence on soil pH and exchangeable Ca and Mg with similar effect sizes. Our results indicate that beech, in comparison to other co‐occurring deciduous tree species, mainly ash, increased acidification at our site. This effect occurred on a small spatial scale and was probably driven by species‐related differences in nutrient cycling via leaf litter. The distribution of beech and ash resulted not only in aboveground diversity of stand structures but also induced a distinct belowground diversity of the soil habitat. 相似文献
12.
Nobuhiro Kaneko Yasunori Sugawara Toshizumi Miyamoto Motohiro Hasegawa Tsutom Hiura 《Pedobiologia》2005,49(6):521-528
Differences in tree species may lead to contrasting soil environments via differences in litter chemical quality and physical environmental factors, such as soil type and soil moisture. However, separating the effects of litter quality and physical environment is difficult under field conditions. Both litter quality and soil environment affect the species composition of the soil animal community. A diversity gradient of canopy tree species (11–25 species) located on homogeneous soil substrate at Tomakomai Experimental Forest of Hokkaido University was used to analyse the relationship between tree species diversity and oribatid mite community structure. Soil samples were collected from three levels of tree species richness (high, intermediate and low) with three replicates each, in July 2000. Leaf area index (LAI) was positively correlated with tree species diversity suggesting higher litter input into the soils with increasing tree diversity. However, the tree species diversity gradient affected neither accumulation of litter on the forest floor nor abundance and species richness of oribatid mites. Canopy and understory plant species richness, LAI, total soil carbon and biomass of epigeic and endogeic earthworms did not significantly affect mite community structure as indicated by redundancy analysis (RDA). The results suggest that oribatid mite community structure is minimally affected by tree species diversity and associated changes in litter diversity. 相似文献
13.
Two key determinants of biological diversity that have been examined in aboveground and aquatic systems are productivity, or resource supply, and physical disturbance. In this study, we examined how these factors interact under field conditions to determine belowground diversity using microarthropods (mites and Collembola) as our test community. To do this, we established a field manipulation experiment consisting of crossed, continuous gradients of nitrogenous (N) fertilizer addition (up to 240 kg N ha?1) and disturbance (imitated trampling by cattle) to produce a gradient of soil nutrient availability and disturbance. Due to the relatively short-term nature of our study (i.e. 2 years), we only detected minimal changes in plant diversity due to the experimental manipulations; in the longer term we would expect to detect changes in plant diversity that could potentially impact on soil fauna. However, disturbance reduced, and additions of N increased, aboveground biomass, reflecting the potential effects of these manipulations on resource availability for soil fauna. We found that disturbance strongly reduced the abundance, diversity, and species richness of oribatid mites and Collembola, but had little effect on predatory mites (Mesostigmata). In contrast, N addition, and therefore resource availability, had little effect on microarthropod community structure, but did increase mesostigmatan mite richness and collembolan abundance at high levels of disturbance. Oribatid community structure was mostly influenced by disturbance, whereas collembolan and mesostigmatan diversity were responsive to N addition, suggesting bottom-up control. That maximal species richness of microarthropod groups overall occurred in undisturbed plots, suggests that the microarthropod community was negatively affected by disturbance. We found no change in microarthropod species richness with high N additions, where plant productivity was greatest, indicating that soil biotic communities are unlikely to be strongly regulated by competition. We conclude that the diversity of soil animals is best explained as a combination of their many varied life history tactics, phenology and the heterogeneity of soils that enable so many species to co-exist. 相似文献
14.
The soil- and litter-dwelling Collembola under four agroforestry tree species (Treculia africana, Dactyladenia (Acioa) barteri, Gliricidia sepium and Leucaena leucocephala) were monitored monthly for a period of 12 months and results were compared with those of a secondary forest and a grass
plot. Treculia and Dactyladenia produced lower quality litter, leading to lower soil temperature and higher soil moisture under those species, relative to
Gliricidia and Leucaena. The agroforestry tree plots contained more soil- and litter-dwelling Collembola than the secondary forest and grass plots.
The slowly decomposing litter under Treculia and Dactyladenia supported more litter Collembola than the quickly decomposing litter under Gliricidia and Leucaena. Soil moisture and temperature were, respectively, positively and negatively correlated with populations of soil Collembola.
Based on the results of this study, it is suggested that the use of prunings of agroforestry tree species as mulch in agroecosystems
would encourage the growth of Collembola populations and enhance their role in decomposition processes.
Received: 28 May 1997 相似文献
15.
Ecosystems of northern North America existed without earthworm fauna until European settlers arrived and introduced European species. The current extent of invasion by some of these species, Lumbricus terrestris L., Octolasion tyrtaeum Savigny and Dendrobaena octaedra Savigny, into an aspen forest in the Canadian Rocky Mountains and the effects of the invasion on soil chemistry, microflora, soil microarthropods and vegetation were investigated. Densities of earthworm species, soil structure, plant coverage and abundance were determined along three transects starting at the edge of the forest. At locations with L. terrestris, litter was incorporated into the soil, and where O. tyrtaeum was present, organic layers were mixed with mineral soil layers. Organic layers disappeared almost entirely when both species occurred together. Carbon and nitrogen concentrations were reduced in organic layers in the presence of L. terrestris and O. tyrtaeum. Microbial biomass and basal respiration were reduced when L. terrestris and O. tyrtaeum were present, presumably due to resource competition and habitat destruction. Microarthropod densities and the number of microarthropod species were strongly reduced in the presence of O. tyrtaeum (−75% and −22%, respectively), probably through mechanical disturbances, increasing compactness of the soil and resource competition. The coverage of some plant species was correlated with earthworm abundance, but the coverage of others was not. Despite harsh climatic conditions, the invasion of boreal forest ecosystems by mineral soil dwelling earthworm species is proceeding and strongly impacts soil structure, soil chemistry, microorganisms, soil microarthropods and vegetation. 相似文献
16.
We examined patterns of microarthropod abundance in oak-hickory (Quercus-Carya) forest ecosystems in southern Ohio (USA) in relation to landscape position and fire frequency. Abundances of various suborders of Acari and Collembola were determined in samples taken in June 1999 in three forested watersheds, one that had been burned annually for four consecutive springs (1996-1999), one that had been burned periodically (1996 and 1999), and an unburned control. Microarthropod abundance was significantly lower in the annually burned watershed than the periodically burned and control watersheds. Since both the periodically burned and annually burned watersheds were burned in April 1999, the lower microarthropod abundance in the annually burned watershed was not simply an immediate effect of burning. At the landscape scale, the abundance of oribatid mites was greater in xeric than intermediate or mesic landscape positions. Within any single watershed, there was no significant linear relationship between litter mass and microarthropod abundance. However, when all three watersheds were combined, there was a significant, positive relationship between litter mass and microarthropod abundance, mainly due to the annually burned watershed where there was very low litter mass and low microarthropod abundance. Both fire frequency and landscape position have significant effects on microarthropod abundance; however, those effects cannot be robustly predicted based solely on forest floor litter mass differences. 相似文献
17.
Astrid Rita Taylor Anne Pflug Dagmar Schröter Volkmar Wolters 《European Journal of Soil Biology》2010,46(2):80-86
The effect of soil organisms on ecosystem processes strongly depends on the composition of the overall community. Community composition however undergoes constant shifts due to pronounced spatio-temporal patterns in biomass and abundance of individual fauna groups. On this background the present experiment aimed to assess the potential impacts of shifts in the biomass of a dominant soil fauna group (microarthropods) on total community composition and on ecosystem processes mediated by fauna or microbes (e.g. decomposition, nitrogen mineralization).Microcosms, filled with spruce litter, hosted soil fauna communities that either contained ambient microarthropod biomass (control) or two elevated levels of microarthropod biomass (1.5 and 2 fold increase), while initial microbial biomass and that of other faunal groups remained unaltered. After an incubation period of 2 months, the biomass of microorganisms and fauna groups as well as ecosystem process variables were determined.The increase in microarthropod biomass at the investigated levels induced changes in the faunal community; mainly via negative or positive feeding interactions between microarthropods and the affected animal groups (Enchytraeidae, Nematoda). The abundance and activity of microorganisms at lower trophic levels however remained unaffected by these alterations; buffering the effect of shifts in the community structure on ecosystem processes. 相似文献
18.
Kam-Rigne Laossi Sbastien Barot Deurival Carvalho Thierry Desjardins Patrick Lavelle Marlucia Martins Danielle Mitja Ana Carolina Rendeiro Guillaume Rousseau Max Sarrazin Elena Velasquez Michel Grimaldi 《Pedobiologia》2008,51(5-6):397-407
We examined the effect of plant diversity on plant production and soil macrofauna density and diversity. Four plants species (Arachis pintoi, an herbaceous legume; Brachiaria brizantha, a perennial grass; Leucaena leucocephala, a legume shrub; Solanum rugosum, a non-legume shrub) were used in a field experiment and communities of all combinations of one, two, three or four species were established. Plant diversity neither significantly affected density and diversity of soil macrofauna nor total plant biomass, however, the biomass of specific plants was negatively affected by plant diversity. Earthworm and ant densities were significantly higher in the presence of A. pintoi although this plant influenced neither the density of the other group nor fauna diversity. Earthworm and diplopod densities increased significantly with shoot biomass of A. pintoi. Fauna diversity increased significantly with shoot biomass (specific and total). Root biomass did not affect fauna density and diversity. Our results suggest that fauna density is affected by litter quality and that it is more affected by resource quantity than quality. Our results also confirm the importance of nitrogen fixers to ecosystem function. 相似文献
19.
H. Zheng Z. Y. Ouyang X. K. Wang H. Miao T. Q. Zhao T. B. Peng 《Land Degradation \u0026amp; Development》2005,16(4):387-396
Significant differences were found in soil physicochemical and biological effects between various forest restoration approaches of a hilly red‐soil region, southern China. Soil quality was the highest in natural secondary forest (0·95), while in sites revegetated with tea‐oil camellia (Camellia oleifera), Chinese fir (Cunninghamia lanceolata) and slash pine (Pinus elliottii) and control sites, integrated soil quality indices were 0·68, 0·55, 0·36 and 0·04 respectively. The lower soil quality of plantations and controls resulted from increased disruption of soil physical structure, lower quality litter fall, lower litter fall production, reduction in microbial biomass, decline of microbial function and loss of soil nutrients due to periodically artificial tending and accelerated soil erosion. Improvement in soil physicochemical properties and enhanced soil microbial function at a natural secondary forest site demonstrated the inherent restoration of these soils. Hence, natural restoration, as well as reducing human disturbance, is a better approach to improving soil properties than returning farmland to planted woodlands. Copyright © 2005 John Wiley & Sons, Ltd. 相似文献
20.
We designed a field experiment to evaluate how restriction of soil faunal movements affects decomposer community structure,
food web architecture, and decomposition of organic matter. Intact soil cores (3cm thick, diameter 16cm) were placed either
in “open” (mesh size 1mm, allowing all meso- and microfauna to move through) or “closed” (27μm, animal movement prevented
except for the smallest microfauna) mesh bags in early May. Before being buried in the forest floor of a mixed spruce stand,
hay litter was placed in the mesh bags in separate litter bags. The samplings took place 2 and 6 months after establishing
the experiment. Additional “field samples” were taken from the adjacent soil to determine possible side effects of the mesh-bags.
Physicochemical conditions, decomposition rate of hay litter, and total respiration of soil cores were identical in the two
bag treatments. Enchytraeids increased significantly in the closed treatment, while macrofauna, such Coleoptera larvae and
dipteran larvae, went close to extinction in the closed bags. The elevated enchytraeid number is in accordance with the findings
of closed microcosm studies, and is best explained by reduced predation by macrofauna. Although a set of 14 mite taxa was
found to distinctively reflect the degree of isolation, neither the total number of individuals nor the number of microarthropod
taxa differed between the bag treatments, or between the bags and the field samples. It is concluded that in the time-span
of one growing season, reduction in the spatial scale does not necessarily reduce the diversity of fauna but can significantly
change the decomposer food-web architecture.
Received: 5 February 1997 相似文献