Collembolan trophic preferences determined using fatty acid distributions and compound-specific stable carbon isotope values

The trophic preferences of soil invertebrates such as Collembola are often determined by the analysis of gut contents, or through visual observations of the location of individuals. As an alternative approach, two species of Collembola, Folsomia candida and Proisotoma minuta, were offered a choice of the soil fungus Cladosporium cladosporioides or the bacterial feeding nematode Panagrellus redivivus; each exhibited distinct fatty acid profiles and stable carbon isotopic compositions. Over 21 days, the fatty acids i15:0, i17:0, 18:1(n-7) and 18:2(n-6) all increased in abundance in both collembolan species consistent with direct routing from the nematode dietary choice which contained a high concentration of these components. Collembolan fatty acid δ¹³C values increased by between 5.7 and 21.6‰ over 21 days reflecting those of the nematode diet. Therefore, both fatty acid profiles and δ¹³C values were consistent with a strong feeding preference of F. candida and P. minuta for the nematodes over the offered fungi. In fact, neither collembolan species consumed any detectable amount of C. cladosporioides. Comparison of the δ¹³C values of the 16:0 and 18:0 fatty acids (which are biosynthesised by the Collembola as well as directly incorporated from the diet) and the 16:1(n-7) and 18:2(n-6) components (which are not biosynthesised by the Collembola) demonstrated that the input of distinct pools of C can lead to large shifts in δ¹³C values between diet and consumer. The fatty acids that were not biosynthesised by Collembola better reflected the δ¹³C values of the diet helping to differentiate between biosynthesised and directly incorporated compounds; an important prerequisite in the interpretation of compound-specific δ¹³C values in trophic behaviour tests. The combination of fatty acid distributions and δ¹³C values is a significant improvement on traditional methods of examining feeding preferences, since it determines directly the assimilated dietary carbon rather than relying on indirect observations, such as the proximity of individuals to a defined food source.     

Translocation of surface litter carbon into soil by Collembola

Soil invertebrates are important in nutrient cycling in soils, but the degree to which mesofauna such as Collembola are responsible for the direct movement of carbon (C) from the litter layer into soil has not yet been ascertained. We used naturally occurring stable C isotopic differences between a C₄ soil and alder leaves (C₃) to examine the effect of the collembolan Folsomia candida on C translocation into soil in laboratory microcosms. Collembolan numbers greatly increased in the presence of alder, but despite large collembolan populations there were no changes in decomposition rate (measured as litter mass loss, cumulative respired CO₂ and alder C:N ratios). Small changes in the δ¹³C values of bulk soil organic matter were detected, but could not be assigned to collembolan activity. However, mean δ¹³C values of soil microbial phospholipid fatty acids (PLFAs) were significantly lower in the presence of alder and Collembola together, demonstrating that collembolan activities resulted in greater availability of litter-derived C to the soil microbial community. Additionally, the presence of Collembola resulted in the translocation of alder-derived compounds (chlorophyll and its breakdown product pheophytin) into soil, demonstrating that Collembola modify soil organic matter at the molecular level. These results are consistent with deposition of collembolan faeces in underlying soil and demonstrate that despite their small size, Collembola contribute directly to C transport in the litter-soil environment.     

Contrasting effects of nitrogen limitation and amino acid imbalance on carbon and nitrogen turnover in three species of Collembola

Soil animal detritivores play an important role in facilitating decomposition processes but little information is available on how the quality of dietary resources affects their stoichiometry of carbon (C) nitrogen (N) and phosphorus (P), and turnover of C and N. This study investigated how a fungal diet, Fusarium culmorum, with a low N content and imbalanced amino acid (AA) composition affected the physiology of three soil-dwelling collembolans (Folsomia candida, Protaphorura fimata and Proisotoma minuta) in comparison to a control diet, Saccharomyces cerevisiae, with a high N content and balanced AA composition. We compared the elemental composition of animals, their growth rates and tissue replacement of C and N. We also measured the individual AA δ¹³C to investigate the extent that Collembola may rely on endogenous sources to compensate for scarcity of essential AAs. The results showed that animal's N content tracked closely the composition of their diets, decreasing from around 10 to 7% N from the high to low N diet. They also had a significant increase of C and a decrease of P. P. fimata was less affected than F. candida and P. minuta. The total incorporation of C and N in the animals due to growth and tissue replacement decreased from 11-17 to 6-12% DM d⁻¹ on the high and low N diet respectively with P. fimata experiencing the smallest change. Essential AAs δ¹³C did not always match perfectly between Collembola species and their diets; particularly on the low N diet. Isotope patterns of AAs indicate that bacteria may have been the alternative source of essential AAs. While the results of this study cannot be extrapolated directly to the dynamics of Collembola populations in the field, they serve to demonstrate their flexibility in adapting physiologically to the temporal and spatial patchiness of the soil environment.     

Fatty acid composition and change in Collembola fed differing diets: identification of trophic biomarkers

To assess the potential of fatty acid (FA) compositions to act as biomarkers in the soil food web, two species of Collembola, Folsomia candida and Proisotoma minuta, were switched to four possible diets: Cladosporium cladosporioides (a common soil fungus), Panagrellus redivivus (a bacteria feeding nematode), Zea mays (maize) and Alnus glutinosa (alder). The change in FA content of the Collembola was observed over the following 39 days. The four diets produced significant shifts in the FA compositions of the Collembola, with P. redivivus causing the most extreme changes; Collembola fed P. redivivus gained complex FA compositions similar to those of the nematode diet. Changes in the relative abundances of some FAs were found to follow negative exponential curves, as the components either accumulated in, or were removed from, the FA pool in the Collembola; abundance half-lives varied between 0.5 and 22.4 days, indicating that Collembolan FA compositions changed readily with the input of new exogenous components. The results demonstrate that Collembolan FA compositions are influenced by diet, and that the abundances of FAs such as i15:0, i17:0 and 18:1(n-7) may be used as biomarkers of nematode consumption by Collembola. In contrast, the C₂₀ polyunsaturated FAs cannot be used as biomarkers for nematode predation as Collembola possess the ability to biosynthesise high abundances of these compounds when not provided by the diet.     

Trophic shift of stable isotopes and fatty acids in Collembola on bacterial diets

Fatty acid (FA) analysis is a promising tool to study trophic relationships in soil food webs. We determined FA biomarkers to trace bacterial food sources (Bacillus megaterium, Pseudomonas putida, Enterobacter aerogenes) of Collembola (Heteromurus nitidus, Protaphorura fimata, Folsomia candida). In addition, δ¹⁵N, δ¹³C, C/N ratio, body weight and NLFA/PLFA ratio (neutral lipid/phospholipid fatty acids) of Collembola were assessed. These measures indicated that P. putida ranked first, B. megaterium second and E. aerogenes third in food quality. FAs specific for bacteria were found in the NLFAs of the Collembola reflecting the respective bacterial diet. Biomarker FAs for gram-positive bacteria were methyl branched i14:0, i15:0, a15:0 and i17:0. Consumption of gram-negative bacteria was reflected by the cyclic form cy17:0 (E. aerogenes, P. putida) and by 16:1ω5 (P. putida).     

New trophic biomarkers for Collembola reared on algal diets

We used fatty acid (FA) analysis to investigate green algae and cyanobacteria as food sources for Collembola. We studied the effects of food quality on body mass and on neutral lipid (NLFA) and phospholipid (PLFA) fatty acid patterns of Collembola. Folsomia candida, Heteromurus nitidus and Protaphorura fimata were fed with common green algae (Chlorella vulgaris), filamentous soil algae (Klebsormidium flaccidum), cyanobacteria (Nostoc commune) and baker's yeast (Saccharomyces cerevisiae). Body mass of F. candida and H. nitidus was highest when reared on C. vulgaris and S. cerevisiae. P. fimata gained the most weight when fed baker's yeast. K. flaccidum and N. commune as resources resulted to low biomass in all Collembola. The four diets caused significant differences in the NLFA and PLFA composition of Collembola after six weeks of feeding. Two new trophic biomarker FAs indicating algal diets were assigned with 16:3ω3,6,9 and 16:2ω6,9, which were only present in NLFAs of Collembola consuming C. vulgaris and K. flaccidum. The amount of FAs from the ω7 family was high in Collembola lipids with cyanobacteria and yeast as food sources, whereas only trace amounts occurred in the NLFA fraction with algae as the resource. In summary, common soil algae and cyanobacteria differed in food quality for Collembola, depending on their growth form (unicellular versus filamentous) and/or secondary metabolites (e.g. cyanobacteria). The new FA biomarkers detected will allow further investigation of these trophic interactions under field conditions; for example, assessing the role of collembolan grazers in the formation of biological soil crusts.     

Feeding guilds in Collembola based on nitrogen stable isotope ratios

In soil a high number of species co-exist without extensive niche differentiation, which was assigned as ‘the enigma of soil animal species diversity’. In particular, the detritivores are regarded as food generalists. We have investigated nitrogen stable isotope ratios (¹⁵N/¹⁴N) of a major decomposer group, the Collembola, to evaluate trophic relationship and determine feeding guilds. Additionally, the δ¹⁵N values of potential food sources such as mosses, lichens and other plant derived material (bark, nuts, leaves) were analysed. The natural variation in nitrogen isotopes was assessed in 20 Collembola taxa from three deciduous forest stands. The δ¹⁵N signature formed a continuum from phycophages/herbivores to primary and secondary decomposers, reflecting a gradual shift from more detrital to more microbial diets. The δ¹⁵N gradient spanned over 9 δ units, which implies a wide range in food sources used. Assuming a shift in ¹⁵N of about 3 ‰ per trophic level, the results indicate a range of three trophic levels. These variations in ¹⁵N/¹⁴N ratios suggest that trophic niches of Collembola species differ and this likely contributes to Collembola species diversity.     

Effects of temperature and life stage on the fatty acid composition of Collembola

Fatty acid (FA) analysis is used as a promising tool to investigate trophic interactions in soil food webs. The FA profile of neutral lipids in consumers is affected by the diet, and the occurrence and amount of certain FAs can reflect feeding strategies. We investigated the lipid composition of the Collembola Folsomia candida, Heteromurus nitidus and Protaphorura fimata with the fungus Chaetomium globosum as food source. The impact of environmental temperature and life stage was assessed, with special respect to linoleic acid (18:2ω6,9) as a marker FA for fungal feeding. In all Collembola species the ratio of C16/C18 in neutral lipid fatty acids (NLFAs) increased with decreasing temperature. In the NLFAs of F. candida and H. nitidus the Unsaturation Index and the amount of 18:2ω6,9 decreased with temperature, whereas in P. fimata effects were the opposite. The composition of phospholipid fatty acids (PLFAs) differed between species, but was little affected by temperature. The degree of unsaturation in NLFAs increased with the age of Collembola, mainly due to higher amounts of 18:2ω6,9 and a lower proportion of 18:1ω9. The biomarker linoleic acid represented over 20% of FAs in all fungal feeding Collembola. Despite considerable influence of temperature and life stage on its proportion, the amount was always higher than in individuals reared on other diets. This suggests that linoleic acid can serve as marker for fungal feeding independent of such physiological variations in Collembola.     

Fungal toxins affect the fitness and stable isotope fractionation of Collembola

We investigated the effect of the fungal toxin sterigmatocystin on the fitness and stable isotope fractionation of two Collembola species (Folsomia candida and Heteromurus nitidus) feeding on mixed vs. single diets. Four knock out mutants of Aspergillus nidulans with the sterigmatocystin production blocked at different steps along the biosynthetic pathway were combined in mixed diets with either the high quality fungus Cladosporium cladosporioides or the low quality fungus A. nidulans (wildtype). Using fungi labeled with stable isotopes (¹³C and ¹⁵N) we evaluated the incorporation of carbon and nitrogen from individual fungi. We hypothesised that (i) Collembola fitness decreases with the putative toxicity of the fungi (ii) Collembola benefit from ingestion of mixed diets due to toxin dilution and (iii) fractionation of ¹³C and ¹⁵N is more pronounced in more toxic diets. Mixed diets did not uniformly improve fitness. Toxin dilution, however, played an important role in Collembola fitness. The fractionation of ¹³C and ¹⁵N varied with sterigmatocystin mutant strains, and Collembola species often differed from the expected enrichment per trophic level. The results show that fungal toxin production may affect stable isotope fractionation, presumably by altering consumer excretion rates necessary for detoxification.     

Seasonal and age-related changes in the stable isotope composition (15N/14N and 13C/12C) of millipedes and collembolans in a temperate forest soil

Seasonal changes in environmental conditions and biotic interactions are often ignored when using stable isotope analysis for reconstructing the trophic structure of soil communities in temperate ecosystems. In this study, we estimated seasonal and age-related changes in δ¹³C and δ¹⁵N values in three epigeic species of collembolans (Pogonognathellus longicornis, Orchesella flavescens and Isotoma viridis) and two litter-dwelling species of millipedes (Polydesmus denticulatus and Leptoiulus proximus) in deciduous and coniferous forest stands in central Russia. Age-related changes in δ¹³C or δ¹⁵N values were either absent or negligible (within 1‰) in L. proximus, but adult and subadult specimens of P. denticulatus were enriched in ¹⁵N compared to early larval stages. Since the adults of P. denticulatus were generally more enriched in ¹⁵N than adults of L. proximus, they presumably occupy more distinct trophic niches than juveniles do. Age-related changes in isotopic composition were small or absent in collembolans studied. Neither δ¹³C nor δ¹⁵N values of millipedes changed significantly during the vegetation season. In contrast, consistent seasonal changes in δ¹³C and δ¹⁵N values were found in collembolans. Increased δ¹³C values coincided with the period of minimum soil moisture and correlated with a decreased C/N ratio in collembolan tissues. These changes can largely be attributed to the depletion of lipid-rich storage tissues. Seasonal changes in δ¹⁵N values were similar among collembolan species, yet slightly varied between habitats. A general trend of increasing δ¹⁵N values from June to September–October may indicate either a reduced importance of non-vascular plants (algae and lichen) in collembolan diet or variation in the isotopic composition of these plants. Overall, our data show that seasonal variations should be taken into account when estimating the isotopic composition of epigeic collembolans in forest soils.     

Influence of elevated CO2 and GM barley on a soil mesofauna community in a mesocosm test system

We hypothesized that the combined effect of rising levels of atmospheric carbon dioxide (CO₂) and increasing use of genetically modified (GM) crops in agriculture may affect soil food-webs. So we designed a study for the assessment of the effects of elevated CO₂ (eCO₂) concentrations and GM barley on a soil-mesofauna community employing a 2nd tier mesocosm test system. The GM barley, Hordeum vulgare cv. Golden Promise, had a modified content of amino acids and it was compared with three non-GM barley cultivated varieties including the isogenic line. Our mesocosm experiment was conducted in a greenhouse at ambient (aCO₂) and eCO₂ (+80 ppm) levels and included a multispecies assemblage of Collembola, Acari and Enchytraeidae with either a GM or conventional spring barley varieties. To detect food-web changes we added dried maize leaves naturally enriched in δ¹³C and δ¹⁵N relative to the soil substrate. Soil, plants and animals were collected after five and eleven weeks. We found that the eCO₂ concentration did not affect the plant biomass, but the predatory mite and two collembolan species showed significantly lower abundances at eCO₂. The densities of three collembolan species (Folsomia fimetaria, Proisotoma minuta and juveniles of Mesaphorura macrochaeta) was significantly lower in the GM treatment compared to some of the non-GM varieties. F. fimetaria was less abundant in presence of GM barley compared to the cultivated barley variety “Netto” at both CO₂ levels, while the density of P. minuta was significantly reduced with the GM barley compared to variety “Netto” at aCO₂ and the isogenic variety at eCO₂. Maize litter acted as a food source for the community, as it was revealed by δ¹³C values in microarthropods. Microarthropod δ¹³C decreased over time, which indicates a diet change of the species towards carbon derived from barley, due to maize litter decomposition. The industrially produced CO₂ gas also had a role as an isotopic marker, as the different δ¹³C values were reflected in the barley and in the collembolan species. GM barley did not affect δ¹³C and δ¹⁵N values of soil animals indicating that the overall trophic structure of the mesofauna community was not changed compared to the non-GM cultivated varieties. The mesocosm methodology integrating stable isotope analysis demonstrates the potential of the multi-species mesocosm as a tool to detect and track changes in the soil trophic interactions in response to environmental pressures, climate and novel agricultural crops.     

Compound-specific δC and δH analyses of plant and soil organic matter: A preliminary assessment of the effects of vegetation change on ecosystem hydrology

Here we present δ¹³C and δ²H data of long-chained, even-numbered (C₂₇-C₃₁) n-alkanes from C3 (trees) and C4 (grasses) plants and from the corresponding soils from a grassland-woodland vegetation sequence in central Queensland, Australia. Our data show that δ¹³C values of the C4 grassland species were heavier relative to those of C3 tree species from the woodland (Acacia leaves) and woody grassland (Atalaya leaves). However, n-alkanes from the C4 grasses had lighter δ²H values relative to the Acacia leaves, but showed no significant difference in δ²H values when compared with C3 Atalaya leaves. These results differ from those of previous studies, showing that C4 grasses had heavier δ²H values relative to C3 grasses and trees. Those observations have been explained by C4 plants accessing the more evaporation-influenced and isotopically heavier surface water and tree roots sourcing deeper, isotopically lighter soil water (“Two-layered soil-water system”). By comparison, our data suggest that ecosystem changes (vegetation “thickening”) can significantly alter the soil hydrological characteristics. This is shown by the heavier δ²H values in the woodland soil compared with lighter δ²H values in the grassland soil, implying that the recent vegetation change (increased tree biomass) in the woodland had altered soil hydrological conditions. Estimated δ²H values of the source-water for vegetation in the grassland and woodland showed that both trees and grasses in open settings accessed water with lighter δ²H values (avg. −46‰) compared with water accessed by trees in the woodland vegetation (avg. −7‰). These data suggest that in semi-arid environments the “two-layer” soil water concept might not apply. Furthermore, our data indicate that compound-specific δ²H and δ¹³C analyses of n-alkanes from soil organic matter can be used to successfully differentiate between water sources of different vegetation types (grasses versus trees) in natural ecosystems.     

Dietary switching of collembola in grassland soil food webs

Soil food webs are characterised by complex direct and indirect effects among the organisms. Consumption of microorganisms by soil animals is considered as an important factor that contributes to the stability of communities, though cascading effects within the food web can be difficult to detect. In a greenhouse experiment, an addition of a high number the fungal feeding collembola Folsomia quadrioculata was applied to grassland soil food webs in monocultures of three plant species: Plantago lanceolata (forb), Lotus corniculatus (legume) and Holcus lanatus (grass). The abundance of microorganisms, determined as the abundances of phospholipid fatty acids (PLFAs) and the abundances of resident invertebrates, nematodes and collembolans, did not change due to the addition of F. quadrioculata. Trophic positions of collembolans were determined by analyses of natural abundances of ¹⁵N stable isotopes. The use of food resources by microorganisms and collembolans was determined by ¹³C analysis of microbial PLFAs and solid samples of collembolans. δ¹³C values of the resident collembola Folsomia fimetaria were lower in the presence of F. quadrioculata than in the control food webs indicating a use of more depleted ¹³C food resources by F. fimetaria. The δ¹⁵N values of F. fimetaria did not change at the addition of F. quadrioculata thus no change of trophic levels was detected. The switch of F. fimetaria to a different food resource could be due to indirect interactions in the food web as the two collembolan species were positioned on different trophic positions, according to different δ¹⁵N values.     

Carbon stable isotope fractionation and trophic transfer of fatty acids in fungal based soil food chains

Stable isotope analysis has been used as a powerful tool in food web studies in terrestrial ecosystems. In addition the occurrence and abundance of fatty acids may serve as indicator for feeding strategies of soil animals. Here we combine both approaches and investigate the fatty acid composition, δ¹³C values of bulk tissues and individual fatty acids in soil organisms. The fungi Chaetomium globosum and Cladosporium cladosporioides were isotopically labelled by fructose derived from either C₃ or C₄ plants, and the fungal-feeding nematode Aphelenchoides sp. was reared on C. globosum. Fungi and nematodes were used as diet for the Collembolan Protaphorura fimata. The sugar source was fractionated differently by fungal lipid metabolism in a species-specific manner that points to a sensitivity of physiological processing to the non-random distribution of ¹³C/¹²C isotopes in the molecule. As a general trend stearic acid (18:0) was depleted in ¹³C compared to the precursor palmitic acid (16:0), whereas its desaturation to oleic acid (18:1 ω9) favoured the ¹³C-rich substrate.Fatty acid profiles of P. fimata varied due to food source, indicating incorporation of dietary fatty acids into Collembolan tissue. Individuals feeding on fungi had lower amounts in C20 fatty acids, with monoenoic C20 forms not present. This pattern likely separates primary consumers (fungivores) from predators (nematode feeders). The isotopic discrimination in ¹³C for bulk Collembola ranged between −2.6 and 1.4‰ and was dependent on fungal species and C₃/C₄ system, suggesting differences at metabolic branch points and/or isotope discrimination of enzymes. Comparison of δ¹³C values in individual fatty acids between consumer and diet generally showed depletion (i.e. de novo synthesis) or no changes (i.e. dietary routing), but the fractionation was not uniform and affected by the type of ingested food. Fatty acid carbon isotopes were more variable than those of bulk tissues, likely due to both the distrimination by enzymes and the different lipid origin (i.e. neutral or polar fraction).     

Use of C abundance to study short-term pig slurry decomposition in the field

Applying pig slurry (PS) on agricultural soils is a common practice. However, its impact on soil organic C dynamics is not clear. This experiment investigated the use of natural ¹³C abundance to study the short-term C mineralization of anaerobically stored PS under field conditions. Measurements of δ¹³C-CO₂ were made on soil air samples obtained from a bare sandy loam during 22 d following incorporation of either PS alone, PS+barley straw, or barley straw alone; an unamended treatment was used as a control. Slurry C was enriched in ¹³C (−20.0‰) because of the high corn (Zea mays L.) content of the animal diet. This value contrasted with δ¹³C of −28.4‰ for the soil organic matter and of −29.0‰ for the barley straw. A peak of high δ¹³CO₂ values (average of −9.2‰) was observed on the day of PS application and was attributed to the dissociation of PS carbonates when mixed with the relatively acidic soil. After this initial burst, 36% of the evolved CO₂ originated from the decomposing PS. After 22 d of incubation, approx. 20% of the PS-C had been lost as CO₂. This short-term field study did not show any priming effect of PS on the mineralization of straw or native soil C. Due to its heterogeneity, the use of the isotopic composition of the evolved CO₂ for estimating PS decomposition requires precaution either through the use of a specific experimental design involving comparable C3 and C4 treatments, or calculations to account for the presence of ¹³C-enriched inorganic C in the PS.     

Atmospheric CO2 enrichment induces life strategy- and species-specific responses of collembolans in the rhizosphere of sugar beet and winter wheat

We studied atmospheric CO₂ enrichment effects on life form types, species composition, dominance structure and individual density of collembolans under cultivation of sugar beet and winter wheat. The study was part of a long-term CO₂ enrichment field experiment (FACE: Free Air CO₂ Enrichment) at the Federal Agricultural Research Centre (FAL) in Braunschweig (Germany), using isotopically labelled CO₂. The stable C-isotopic signature (δ¹³C) of collembolan species, plant material, and soil indicated CO₂ impacts on C translocation. The δ¹³C values of both crops significantly increased from above-ground to below-ground plant parts and significantly decreased under FACE conditions. The δ¹³C values of collembolan species differed significantly depending on CO₂ treatment and crop and showed a distinct tendency depending on plant growth stage. The extent, to which δ¹³C values of collembolans decreased under FACE conditions, was species- and life strategy-dependent. The stable C-isotopic signatures of euedaphic and hemiedaphic species were similar in the control, but, depending on crop, differently affected by atmospheric CO₂ enrichment. Under winter wheat cultivation, hemiedaphic species showed more negative δ¹³C values than euedaphic ones under FACE conditions. CO₂ enrichment effects on occurrence, density and dominance distribution of the collembolan species differed strongly between crops and their developmental stages, which reveal crop-specific below-ground effects due to different food qualities in the rhizosphere. CO₂ impacts were stronger under sugar beet compared to winter wheat cultivation. Independent of crop, CO₂ enrichment enhanced the diversity of collembolans before harvest and increased the proportion of hemiedaphic in relation to euedaphic species in a community. Our results on collembolan communities imply CO₂-induced changes in the root-derived carbon resources used by the soil food web. The present study reveals atmospheric CO₂ enrichment impacts to specifically affect collembolan species according to their food preferences.     

Three-source partitioning of CO2 efflux from soil planted with maize by C natural abundance fails due to inactive microbial biomass

A theoretical approach to the partitioning of carbon dioxide (CO₂) efflux from soil with a C₃ vegetation history planted with maize (Zea mays), a C₄ plant, into three sources, root respiration (RR), rhizomicrobial respiration (RMR), and microbial soil organic matter (SOM) decomposition (SOMD), was examined. The δ¹³C values of SOM, roots, microbial biomass, and total CO₂ efflux were measured during a 40-day growing period. A three-source isotopic mass balance based on the measured δ¹³C values and on assumptions made in other studies showed that RR, RMR, and SOMD amounted to 91%, 4%, and 5%, respectively. Two assumptions were thoroughly examined in a sensitivity analysis: the absence of ¹³C fractionation and the conformity of δ¹³C of microbial CO₂ and that of microbial biomass. This approach strongly overestimated RR and underestimated RMR and microbial SOMD. CO₂ efflux from unplanted soil was enriched in ¹³C by 2.0‰ compared to microbial biomass. The consideration of this ¹³C fractionation in the mass balance equation changed the proportions of RR and RMR by only 4% and did not affect SOMD. A calculated δ¹³C value of microbial CO₂ by a mass balance equation including active and inactive parts of microbial biomass was used to adjust a hypothetical below-ground CO₂ partitioning to the measured and literature data. The active microbial biomass in the rhizosphere amounted to 37% to achieve an appropriate ratio between RR and RMR compared to measured data. Therefore, the three-source partitioning approach failed due to a low active portion of microbial biomass, which is the main microbial CO₂ source controlling the δ¹³C value of total microbial biomass. Since fumigation-extraction reflects total microbial biomass, its δ¹³C value was unsuitable to predict δ¹³C of released microbial CO₂ after a C₃-C₄ vegetation change. The second adjustment to the CO₂ partitioning results in the literature showed that at least 71% of the active microbial biomass utilizing maize rhizodeposits would be necessary to achieve that proportion between RR and RMR observed by other approaches based on ¹⁴C labelling. The method for partitioning total below-ground CO₂ efflux into three sources using a natural ¹³C labelling technique failed due to the small proportion of active microbial biomass in the rhizosphere. This small active fraction led to a discrepancy between δ¹³C values of microbial biomass and of microbially respired CO₂.     

Consumption of mycorrhizal and saprophytic fungi by Collembola in grassland soils

Although soil-dwelling Collembola can influence plant growth and nutrient cycling, their specific role in soil food webs is poorly understood. Soil-free microcosm studies suggest that Collembola are primarily fungivores where they feed preferentially on saprophytic fungi (SF) over other fungal types. We directly assessed collembolan consumption of arbuscular mycorrhizal fungi (AMF) and SF using plant-soil mesocosms and natural abundance stable carbon isotope techniques. Mycorrhizal Andropogon gerardii (C₄ grass) seedlings were placed in pots containing Collembola and soil from a C₃ plant dominated site, while mycorrhizal Pascopyrum smithii (C₃ grass) seedlings were placed in pots with Collembola and soil collected at a C₄ plant dominated site. After 6 weeks, collembolans assimilated carbon derived from C₃ and C₄ sources in both A. gerardii and P. smithii treatments. Comparing Collembola isotope values in AMF vs. AMF-suppressed treatments, our data show that both AMF and SF were consumed in these experimental soil environments.     

Savanna-derived organic matter remaining in arable soils of the South African Highveld long-term mixed cropping: Evidence from C and N natural abundance

Sustainable agriculture requires the formation of new humus from the crops. We utilized ¹³C and ¹⁵N signatures of soil organic matter to assess how rapidly wheat/maize cropping contributed to the humus formation in coarse-textured savanna soils of the South African Highveld. Composite samples were taken from the top 20 cm of soils (Plinthustalfs) cropped for lengths of time varying from 0 to 98 years, after conversion from native grassland savanna (C4). We performed natural ¹³C and ¹⁵N abundance measurements on bulk and particle-size fractions. The bulk soil δ¹³C values steadily decreased from −14.6 in (C4 dominated) grassland to −16.5‰ after 90 years of arable cropping. This δ¹³C shift was attributable to increasing replacement of savanna-derived C by wheat crop (C3) C which dominated over maize (C4) inputs. After calculating the annual C input from the crop yields and the output from literature data, by using a stepwise C replacement model, we were able to correct the soil δ¹³C data for the irregular maize inputs for a period of about one century. Within 90 years of cropping 41-89% of the remaining soil organic matter was crop-derived in the three studied agroecosystems. The surface soil C stocks after 90 years of the wheat/maize crop rotation could accurately be described with the Rothamsted Carbon Model, but modelled C inputs to the soil were very low. The coarse sand fraction reflected temporal fluctuations in ¹³C of the last C₃ or C₄ cropping and the silt fraction evidenced selective erosion loss of old savanna-derived C. Bulk soil ¹⁵N did not change with increasing cropping length. Decreasing δ¹⁵N values caused by fertilizer N inputs with prolonged arable cropping were only detected for the coarse sand fraction. This indicated that the present N fertilization was not retained in stable soil C pool. Clearly, conventional cropping practices on the South African highlands neither contribute to the preservation of old savanna C and N, nor the effective humus reformation by the crops.     

Natural abundance δN and δC of DNA extracted from soil

We report the first simultaneous measurements of δ¹⁵N and δ¹³C of DNA extracted from surface soils. The isotopic composition of DNA differed significantly among nine different soils. The δ¹³C and δ¹⁵N of DNA was correlated with δ¹³C and δ¹⁵N of soil, respectively, suggesting that the isotopic composition of DNA is strongly influenced by the isotopic composition of soil organic matter. However, in all samples DNA was enriched in ¹³C relative to soil, indicating microorganisms fractionated C during assimilation or preferentially used ¹³C enriched substrates. Enrichment of DNA in ¹⁵N relative to soil was not consistently observed, but there were significant differences between δ¹⁵N of DNA and δ¹⁵N of soil for three different sites, suggesting microorganisms are fractionating N or preferentially using N substrates at different rates across these contrasting ecosystems. There was a strong linear correlation between δ¹⁵N of DNA and δ¹⁵N of the microbial biomass, which indicated DNA was depleted in ¹⁵N relative to the microbial biomass by approximately 3.4‰. Our results show that accurate and precise isotopic measurements of C and N in DNA extracted from the soil are feasible, and that these analyses may provide powerful tools for elucidating C and N cycling processes through soil microorganisms.