Search

Protein hotspots in soil, such as those associated with decaying soil fauna or plant litter, may produce ephemeral patches of disproportionately high soil nutrients. These hotspots may occur at the macro- and microscale in close proximity to plant roots, however, the likely concentration of soluble products produced in these hotspots remains poorly understood. To address this, we buried two contrasting biomass residues in soil, namely earthworm (Lumbricus terrestris) and clover (Trifolium repens). Their transformation to amino acids, NH4+ and NO3− were monitored continually over 6 days using microdialysis. All treatments showed greater soluble nitrogen (N) concentrations compared to the unamended controls. The highest concentrations of both amino acids (12.9 mM after 12 h) and NH4+ (45.3 mM after 6 h) were generated in the vicinity of decomposing earthworm. In comparison, dried clover residues yielded 2.7 mM of amino acids at 6 h. After 12 h, amino acid and NH4+ concentrations in both earthworm and dried clover treatments showed a steep decline, returning close to background levels (

We hypothesized that the rhizosphere priming effect (RPE) of soil organic matter by mutant barley lacking root hairs is dependant on a large network of symbiotic arbuscular mycorrhizal fungi (AMF). We thus predicted that fertilizing with phosphate-P would reduce AMF abundance and, in turn, reduce RPE of mutant barley. We packed microcosms with a P-responsive soil in which we grew mutant barley lacking root hairs as well as wild type barley and narrowleaf plantain, each possessing root hairs. One set of microcosms was fertilized with phosphate-P while another set was not fertilized. The plants were grown in a labelling chamber with 13C-depleted CO2. Soil respiration and δ13C of headspace CO2 were measured after 3, 4 and 5 weeks and RPE was calculated using an isotope mass balance approach. Root hair length was measured and soils were analyzed for the 16:1ω5 neutral lipid fatty acid (i.e. AMF biomarker). AMF abundance was greater, whereas RPE was lower, in mutant barley soil under low-P than under high-P conditions. In the other two plant-types, P had no effect on AMF or on RPE. As our results contradict our prediction, we propose an alternative explanation based on plant N demand under high-P.

Maize root traits associated with soil organic matter (SOM) mineralization were demonstrated to have a heritable genetic basis. We show root length, root diameter and cumulative root-derived C mineralization to be strong predictors of SOM-C mineralization and identify two candidate genes associated with enhanced SOM-C mineralization rates. There is potential to target these genes to enhance release of nutrients from SOM to support crop nutrition within sustainable maize cropping systems.

Mechanisms controlling carbon stabilisation in soil and its feedback to climate change are of considerable importance. Microbial substrate-use efficiency is an important property during decomposition of soil organic matter. It determines the allocation of substrate towards biosynthetic stabilisation of carbon and for respiratory losses into the atmosphere. Previously, it was observed that substrate-use efficiency declines with an increase in temperature and that it varies across organic substrates. Yet, our mechanistic understanding of processes causing the temperature sensitivity of substrate-use efficiency is limited. Changes in substrate-use efficiency could be triggered by (i) shifts in the active components of microbial communities, (ii) changes in microbial physiology within the same community, or (iii) a combination of both. In the present study, we evaluated the link between microbial community composition and substrate-use efficiency, combining measurements of carbon mineralisation and microbial energetics. We found only minor shifts in microbial community composition, despite large differences in substrate-use efficiencies across incubation temperatures and substrate additions. We conclude that short-term changes in substrate-use efficiency were mainly caused by changes in microbial physiology, but emphasize that future studies should focus on resolving long-term trade-offs between physiological and community influences on substrate-use efficiency.

Rhizodeposition is an important source of substrate for microbial communities, supporting activities including soil organic matter (SOM) and nutrient cycling. Therefore, it is a potential trait of interest for crop plants, particularly in the context of variety selection for sustainable production systems. However, we do not have a good understanding of (i) whether there is significant variation in root-C deposition between varieties of important agricultural crops and (ii) whether variation in C deposition between varieties leads to major differences in C cycling in soil. In two experiments, we assessed variations in C deposition amongst barley genotypes and their respective impacts on microbial activity and SOM dynamics. In experiment 1, we applied 13C–CO2 labelling to selected barley recombinant chromosome substitution lines (RCSLs) and traced root-derived C in surface soil CO2 efflux, soil microbial biomass-C (MBC), soil solution, and soil particle-size fractions. In experiment 2, we conducted MicroResp analysis using 15 ecologically relevant C substrates to assess the impacts of barley genotypes on microbial activity. Soil respiration measurements (partitioned into plant- and SOM-derived components) revealed genotype-specific effects on plant-derived C, SOM-derived C and total C respired as CO2. For particle-size fractionation, we found that incorporation of plant-derived C to the silt-and-clay fraction varied between genotypes, indicating differences in relative stabilization of root-derived C as a result of barley genotype. Our data did not indicate genotype effects on total MBC size or dissolved organic-C (DOC) in soil solutions, but significant differences in plant-derived MBC and DOC were observed. MicroResp analysis showed differential utilization of 7 substrates (glucose, trehalose, lignin, arabinose, alanine, aminobutyric acid and lysine) revealing variation in community level physiological profiles (CLPPs) of soil microbes as impacted by barley genotypes. Furthermore, we found significant clustering of microbial CLPPs as a function of RCSLs and parent lines (Caesarea 26-24 and Harrington) suggesting a strong plant genetic control of the barley microbiome, and that this genetic control is heritable. Our results demonstrate barley genotype-specific effects on soil processes, revealing the potential for germplasm selection and variety improvement in barley to support sustainable production systems.

Soil organic matter (SOM) is the dominant store of nutrients required for plant growth, but the availability of these nutrients is dependent on transformations mediated by the microbial biomass. The addition of labile C to soil is known to alter SOM turnover (priming effect, PE), but understanding of this is limited, particularly with respect to impact on gross nitrogen (N) fluxes. Here we examined relationships between C and N fluxes from SOM under primed and non-primed conditions in two soils. Stable isotopes (13C and 15N) were used to measure gross C and N fluxes from SOM and to differentiate between SOM mineralised due to priming and that from basal mineralisation. 13C-glucose was added daily to simulate the effect of addition of labile C on SOM-C and –N mineralisation within the rhizosphere. Addition of glucose increased both gross N and C mineralisation from SOM. However, the C-to-N ratio of the mineralised flux from ‘primed’ SOM was 5:1, whereas the C-to-N ratio of the basal mineralised flux was 20:1 indicating that priming acted on specific organic matter pools. This result is consistent with the concept that priming is a distinct N-mining response of the microbial biomass, as opposed to an acceleration of the basal flux. Our data suggest that C and N fluxes are not directly linked through their gross stoichiometry in SOM. This is due to the heterogeneity and overall passiveness of OM relative to the dynamic nature of mineralisation fluxes and source pools, and in primed systems the mineralisation of N-rich compounds.

Temperate forest soils are important carbon (C) sinks, where the C-stock is largely determined by the balance of leaf inputs and losses through respiration. However, studies dealing with leaf inputs to coniferous forest soils are limited although coniferous forests are widespread through the Northern temperate zone. In this study, we focused on the effects of soluble, insoluble and whole-tissue coniferous needle fractions on soil organic matter (SOM) decomposition and C storage in soil fractions. In addition, the effect of future increased C input was tested by applying a doubled amount of the soluble fraction (whole-tissue + soluble fraction). 13C-labelled needles were produced from spruce seedlings in growth chambers and needle fractions were added to the coniferous forest soil in laboratory microcosms. CO2 respired during incubation from the microcosms was partitioned into needle- and SOM-derived components. After seven months, soils were destructively harvested and analysed for C content in soil fractions and microbial community composition. The soluble, insoluble and whole-tissue fractions resulted in cumulative priming (increased SOM-derived CO2 relative to unamended controls) of 25 ± 8%, 40 ± 1%, and 39 ± 7%, respectively. The doubled soluble-C addition caused a slightly lower priming (38 ± 2%) than the whole-tissue fraction alone. The addition of needle fractions did not significantly affect the C content of soil fractions. However, the soluble fraction retained in soil was mainly found adsorbed onto mineral particles, whereas the insoluble and whole-tissue fractions occurred mainly as free particulate organic matter or adsorbed onto mineral particles. The insoluble and whole-tissue fraction led to increased fungal abundance and decreased abundance of G+ bacteria and actinobacteria. All the fractions were primarily incorporated into fungal biomass after seven months suggesting that fungi were the main consumers of all needle fractions after the labile C had been depleted. When considering all the C gains and losses, the addition of all needle fractions resulted in net soil C increase. This suggests that, although the input of the coniferous needles leads to some C losses through the priming of SOM decomposition, these C losses are compensated by new C storage either in SOM fractions or microbial biomass.

Carbon exchange across the soil–atmosphere interface has major implications for environmental change, but a consensus on the effect of temperature on soil organic matter (SOM) mineralisation remains elusive. In this study we investigated the temperature sensitivities of basal respiration (partitioned into recent and older SOM sources) and of additional SOM mineralisation associated with the addition of labile C to soil (priming effects). Soil samples were incubated at 15, 20, 25 and 30 °C, and following a 14 day stabilisation period, daily amendments of 13C-enriched glucose (0.2 mg g−1 soil) were applied (control treatments received water only). Soils were collected where C4 maize had recently been cultivated across fields which were historically planted with C3 spring barley. This enabled basal SOM mineralisation to be partitioned into recent (

Terrestrial carbon (C) storage and turnover are of major interest under changing climatic conditions. We present a laboratory microcosm study investigating the effects of anticipated climatic conditions on the soil microbial community and related changes in soil organic matter (SOM) decomposition. Soil samples were taken from a heath-land after six years of exposure to elevated carbon dioxide (eCO 2 ) in combination with summer drought (D) and increased temperature (T). Soil C-dynamics were investigated in soils from: (i) ambient, (ii) eCO 2 , and (iii) plots exposed to the combination of factors simulating future climatic conditions (TDeCO 2 ) that simulate conditions predicted for Denmark in 2075. 13 C enriched glucose (3 atom% excess) was added to soil microcosms, soil CO 2 efflux was measured over a period of two weeks and separated into glucose- and SOM-derived C. Microbial biomass was measured using chloroform fumigation extraction, and compound-specific phospholipid fatty acid analysis was used to determine microbial community composition and substrate use. We observed that glucose additions induced SOM priming in ambient and eCO 2 treated soils, but not in soil exposed to future climatic conditions. Climate treatments and glucose additions did not affect relative abundances of microbial functional groups but the fate of glucose through the microbial community was changed by climate treatments as revealed by the incorporation of 13 C in PLFAs. Soil treated with eCO 2 showed a high flow of glucose through gram-positive bacteria whereas in ambient and future soils utilization of glucose by actinomycetes and fungi (putative SOM-decomposers) was greater. Our results suggest that individual climate change factors may influence pathways of C-flux through microbial communities and therefore affect soil processes; these factors may counterbalance each other and maintain ecosystem stability. This highlights the importance of studying climate change factors in combination to fully assess consequences of environmental change on plant–soil systems.

Quantifying the potential for eroding agricultural soils to act as sinks or sources of atmospheric carbon relies on accounting for the pools and fluxes of soil organic carbon (SOC) and nutrients, e.g. nitrogen (N), affected by erosion. Herein, we report the outcomes of an experiment where a C 4 maize ( Zea mays ) crop ( δ 13 C = −12.1‰) was cultivated and incorporated for 2 years to introduce a ‘pulse’ of 13 C-enriched SOC to a C 3 arable soil ( δ 13 C = −27.4‰). Soils were sampled at eroding (top slope and upper slope) and depositional (lower slope and slope foot) positions of an accelerated erosion pathway that were confirmed using 137 Cs measurements. The sand particle-sized fraction (63–2000 μm) was predominant and increased in the depositional slope positions due to selective loss of fine particles and preferential deposition of the coarsest fraction of transported sediment. There was a significant isometric relationship between the percentage SOC and total N: top slope > upper slope > lower slope, with similar values in the slope foot to the top slope. The δ 15 N values of the soils were enriched (7.3‰) at the slope foot, compared with the other slope positions (average 6.3‰), suggesting increased denitrification rates. The δ 13 C values of the soil microbial biomass C extracted from surface soils (0–5 cm) at each slope position showed that the proportion of maize C being incorporated into the soil microbial biomass declined in the downslope direction from 54% (top slope) to 43% (upper slope) to 18% (lower slope) in inverse proportion to the size of the soil microbial biomass, and increased to 41% at the slope foot. This suggests dynamic replacement of the SOC with crop C in the eroding slope positions and dilution of the transported C by C3-SOC in the depositional slope positions. This paper is evidence that erosional distribution of soil carbon leads to differential microbial utilisation of SOC between eroding and depositional sites.

The overall processes by which carbon is fixed by plants in photosynthesis then released into the soil by rhizodeposition and subsequently utilized by soil micro-organisms, links the atmospheric and soil carbon pools. The objective of this study was to determine the plant derived 13C incorporated into the phospholipid fatty acid (PLFA) pattern in paddy soil, to test whether utilization of rice rhizodeposition carbon by soil micro-organisms is affected by soil water status. This is essential to understand the importance of flooded conditions in regulating soil microbial community structure and activity in wetland rice systems. Rice plants were grown in soil derived from a paddy system under controlled irrigation (CI), or with continuous waterlogging (CW). Most of the 13C-labelled rice rhizodeposition carbon was distributed into the PLFAs 16:0, 18:1ω7 and 18:1ω9 in both the CW and CI treatments. The bacterial PLFAs i15:0 and a15:0, both indicative of gram positive bacteria, were relatively more abundant in the treatments without rice plants. When rice plants were present rates of 13C-incorporation into i15:0 and a15:0 was slow; the microbes containing these PLFAs may derive most of their carbon from more recalcitrant C (soil organic matter). PLFAs, 18:1ω7 and 16:1ω7c, indicative of gram negative bacteria showed a greater amount incorporation of labelled plant derived carbon in the CW treatment. In contrast, 18:2ω6,9 indicative of fungi and 18:1ω9 indicative of aerobes but also potentially fungi and plant roots had greater incorporation in the CI treatment. The greater root mass concomitant with lower incorporation of 13C into the total PLFA pool in the CW treatment suggests that the microbial communities in wetland rice soil are limited by factors other than substrate availability in flooded conditions. In this study differing soil microbial communities were established through manipulating the water status of paddy soils. Steady state 13C labelling enabled us to determine that the microbial community utilizing plant derived carbon was also affected by water status.

Microbial communities in soil are highly species-rich, recognition of which has led to the view that functional redundancy within communities may buffer many impacts of altered community structure on soil functions. In this study we investigated the impact of long-term (>50 years) exclusion of plant-inputs (bare-fallow treatment) on soil microbial community structure and on the ability of the microbial biomass to mineralise tracer additions of 13 C-labelled plant-derived C-substrates. Exclusion of plant-inputs resulted in depletion of soil organic matter (SOM) and a reduction in microbial biomass size. The microbial community structure was also strongly affected, as indicated by the distinct phospholipid fatty acid (PLFA) profiles in bare-fallow and grassland soils. Mineralisation of labile plant-derived substrates was not perturbed by the bare-fallow treatment. The incorporation of labile plant-derived C into PLFA biomarkers was found to differ between soils, reflecting the distinct community structures of the soils and indicating that these substrates were utilised by a broad range of microbial groups. In contrast, the mineralisation of recalcitrant plant-derived substrates was reduced in bare-fallow soil and the fate of substrate-derived C within PLFA biomarkers was, initially, similar between the soils. These results indicate that utilisation of these recalcitrant substrates was a function restricted to specific groups, and that exclusion of plant-derived inputs to soil had reduced the capacity of bare-fallow microbial communities to utilise this substrate type. Therefore, the study suggests that long-term selective pressure on microbial communities, resulting in altered community structure, may also result in altered functional attributes. This structure–function relationship was apparent for utilisation of recalcitrant plant-derived substrates, but not for the more widely distributed attribute of labile C-substrate utilisation.

Altered rates of native soil organic matter (SOM) mineralisation in the presence of labile C substrate (‘priming’), is increasingly recognised as central to the coupling of plant and soil-biological productivity and potentially as a key process mediating the C-balance of soils. However, the mechanisms and controls of SOM-priming are not well understood. In this study we manipulated microbial biomass size and composition (chloroform fumigation) and mineral nutrient availability to investigate controls of SOM-priming. Effects of applied substrate ( 13 C-glucose) on mineralisation of native SOM were quantified by isotopic partitioning of soil respiration. In addition, the respective contributions of SOM-C and substrate-derived C to microbial biomass carbon (MBC) were quantified to account for pool-substitution effects (‘apparent priming’). Phospholipid fatty acid (PLFA) profiles of the soils were determined to establish treatment effects on microbial community structure, while the 13 C-enrichment of PLFA biomarkers was used to establish pathways of substrate-derived C-flux through the microbial communities. The results indicated that glucose additions increased SOM-mineralisation in all treatments (positive priming). The magnitude of priming was reduced in fumigated soils, concurrent with reduced substrate-derived C-flux through putative SOM-mineralising organisms (fungi and actinomycetes). Nutrient additions reduced the magnitude of positive priming in non-fumigated soils, but did not affect the distribution of substrate-derived C in microbial communities. The results support the view that microbial community composition is a determinant of SOM-mineralisation, with evidence that utilisation of labile substrate by fungal and actinomycete (but not Gram-negative) populations promotes positive SOM-priming.

Plants link atmospheric and soil carbon pools through CO2 fixation, carbon translocation, respiration and rhizodeposition. Within soil, microbial communities both mediate carbon-sequestration and return to the atmosphere through respiration. The balance of microbial use of plant-derived and soil organic matter (SOM) carbon sources and the influence of plant-derived inputs on microbial activity are key determinants of soil carbon-balance, but are difficult to quantify. In this study we applied continuous 13C-labelling to soil-grown Lolium perenne, imposing atmospheric CO2 concentrations and nutrient additions as experimental treatments. The relative use of plant- and SOM-carbon by microbial communities was quantified by compound-specific 13C-analysis of phospholipid fatty acids (PLFAs). An isotopic mass-balance approach was applied to partition the substrate sources to soil respiration (i.e. plant- and SOM-derived), allowing direct quantification of SOM-mineralisation. Increased CO2 concentration and nutrient amendment each increased plant growth and rhizodeposition, but did not greatly alter microbial substrate use in soil. However, the increased root growth and rhizosphere volume with elevated CO2 and nutrient amendment resulted in increased rates of SOM-mineralisation per experimental unit. As rhizosphere microbial communities utilise both plant- and SOM C-sources, the results demonstrate that plant-induced priming of SOM-mineralisation can be driven by factors increasing plant growth. That the balance of microbial C-use was not affected on a specific basis may suggest that the treatments did not affect soil C-balance in this study.

Plant inputs of organic material to soil are thought to be key determinants of microbial activity, community composition and processes. However, the identity of organisms utilising these chemically diverse inputs is not well understood. In this study, we applied tracer amounts of highly enriched, 13C-labelled plant tissue fractions (whole, insoluble and soluble) to soil cores that either allowed or prevented access to roots and mycorrhizal fungi. For all tissue fractions, C derived from the additions was detected rapidly (

The influence of repeated defoliation on soil microbial community (SMC) structure and root turnover was assessed in two contrasting Lolium perenne cultivars (AberDove and S23) grown in fertilised (+F) and non-fertilised (NF) soil. BiOLOG sole carbon source utilisation profiles (SCSUPs) indicated consistently greater potential carbon utilisation in defoliated (+D) compared to non-defoliated (ND) soils regardless of cultivar and fertiliser, and was accounted for in a variety of substrate groups (sugars, carboxylic, amino and phenolic acids). Potential carbon utilisation was also stimulated in +F compared to NF soils, primarily through increased potential utilisation of carboxylic acids. PLFA indicators for the bacterial biomass did not significantly differ between cultivar, soil fertilisation, or defoliation. Defoliated swards grown in fertilised soil (+F+D) had a higher fungal:bacterial ratio and a greater bacterial stress index (cy19:0/18:1w7c), compared to that of +F ND, NF ND and NF+D, and regardless of cultivar. Overall SMC structure (canonical variate (CV) analysis of PLFAs) discriminated based on cultivar, defoliation and soil fertilisation. Primary discrimination of the SMCs could be related to differences in root density and total plant biomass, and in the case of NF soils, secondary community shifts, evident with defoliation, related to root disappearance over the growing season. Despite the strong common effects of defoliation, and to a lesser extent soil fertilisation, cultivar specific drivers of the soil microbial community were maintained, resulting in consistent, but subtle, discrimination of the SMC associated with the contrasting L. perenne cultivars.

Intra-species variation in response to defoliation and soil amendment has been largely neglected in terms of the soil microbial community (SMC). The influence of defoliation and soil fertiliser amendment on the structure of the SMC was assessed with two Lolium perenne cultivars contrasting in ability to accumulate storage reserves. Plant response to defoliation was cultivar specific and depended on the nutrient amendment of the soil. Results suggested a greater ability to alter plant biomass allocation in the low carbohydrate accumulating cultivar (S23) compared to the high carbohydrate cultivar (AberDove) when grown in improved (IMP), but not in unimproved (UNI), soil. Although differences in plant growth parameters were evident, no treatment effects were detected in the size of the active microbial biomass (total phospholipid fatty acid (PLFA) 313.8 nmol g−1 soil±33.9) or proportions of PLFA signature groups. A lower average well colour development (AWCD) of Biolog sole carbon source utilisation profiles (SCSUPs) in defoliated (D) compared to non-defoliated (ND) treatments may be indicative of lower root exudation 1 week following defoliation, as a consequence of lower root non-structural carbohydrate (NSC) concentrations. Within the bacterial community the lower cyclopropyl-to-precursor ratio of PLFAs, and the trans/cis ratio of 16:1w7, in UNI relative to IMP soil treatments indicates lower physiological stress in UNI soils regardless of L. perenne cultivar. Discrimination of broad scale SMC structure, measured by PLFA analysis, revealed that soil treatment interacted strongly with cultivar and defoliation. In IMP soils the SMCs discriminated between cultivars while defoliation had little effect. Conversely, in UNI soils defoliation caused a common shift in the SMC associated with both cultivars, causing convergence of overall community structure. Separation of SMC structure along the primary canonical axis correlated most strongly (P

The paper by Kemmitt et al. [2008. Mineralization of native soil organic matter is not regulated by the size, activity or composition of the soil microbial biomass – a new perspective. Soil Biology and Biochemistry 40, 61–73] proposing the existence of an abiotic regulatory gate that controls the rate-limiting step of stabilised soil organic matter (SOM) mineralization, has initiated a fundamental and far-reaching debate. In this contribution the implications of a functioning abiotic regulatory gate are considered in the context of microbial community diversity and soil carbon cycling. I argue that although the evidence presented in support of the regulatory gate is strong, abiotic routes for SOM-mineralization function in parallel with biologically mediated mechanisms. Evidence is now accumulating that, in the presence of plant-inputs to soil, enhanced microbial mobilisation of SOM into biomass is a quantitatively important and ubiquitous process. I argue that this mineralization of SOM is fuelled by energy-rich substrates and is driven by microbial nutrient-demand. This implies that the mineralization of stabilised SOM and the turnover of C-inputs from current vegetation are intimately linked through the functioning of microbial communities associated with plants. This suggests that the microbial ‘eye of the needle’ is a crucial control-point in determining the carbon balance of soils. Fortunately, there are now excellent methods that allow quantification of SOM- and plant-derived C-fluxes through the members of soil microbial communities, and will also allow quantification of the relative importance of the abiotic and biotic routes of SOM-mineralization.

The effects of elevated atmospheric CO 2 concentration on alterations, both qualitatively and quantitatively, of exuded compounds from the roots of Lolium perenne seedlings were investigated by growing plants in a sterilised sand microcosm unit. In addition, the effect of CO 2 treatment on carbon substrate utilisation of microbial populations extracted from the rhizosphere of L. perenne seedlings grown in soil microcosm units was examined and alterations on microbial activity and diversity assessed using a commercially-available redox-based sole C source utilisation test (Biolog®) including additional exudate compounds. Both types of microcosm units (sand and soil) were maintained at specific growth conditions under two CO 2 regimes (450 and 720 μ mol mol −1 ). Growth of L. perenne seedlings from both types of microcosm units was enhanced under elevated atmospheric CO 2 although the root-to-shoot ratios were not significantly altered, indicating no gross change in dry matter partitioning. Cumulative total organic carbon (TOC) release in the exudate material over the duration of the experiment was significantly ( P ≤0.05) higher from ambient-grown seedlings despite a significant ( P ≤0.05) increase in the dry weight of roots of the elevated CO 2 grown seedlings as determined at harvest. Over the individual sampling periods TOC release was significantly ( P ≤0.05) higher from elevated CO 2 grown seedlings on only one occasion (21 d). Qualitative differences, measured between d 1–6 and 14–18, also occurred with elevated CO 2 treatment decreasing the amount of phenolic acids and total sugars at the latter sampling period compared to ambient CO 2 seedlings. Total numbers of bacteria were significantly ( P ≤0.05) decreased under elevated CO 2 although culturable numbers significantly ( P ≤0.05) increased. This increase in culturable microorganisms may explain the faster carbon source utilisation rates of the elevated CO 2 treatment. No change in morphotypes of microbial colonies were observed suggesting a quantitative difference due to elevated CO 2 treatment only.

The structure of microbial communities in the rhizospheres of ryegrass and wheat, growing at an elevated atmospheric CO2 concentration, was investigated using broad-scale DNA techniques. Community DNA hybridisation and %G+C base profiling by thermal denaturation assess changes at the whole microbial community level. DNA analysis of the rhizosphere of ryegrass grown in soil microcosms for 28 or 42 d, showed only minor differences between plants grown at 450 or 720 μl CO2 l−1. In a second experiment with ryegrass, 5 of 10 replicate microcosms were pulse labelled with 14CO2 and 5 simultaneously sampled for DNA analysis. Carbon partitioning below ground showed changes due to the elevated CO2, notably an increased proportion of fixed carbon in non-microbial biomass residue in the rhizosphere. There was again no effect of elevated CO2 on rhizosphere microbial community structure. Community DNA hybridisation indicated that the rhizosphere communities under ambient and elevated CO2 were 86% similar (unlikely to be a biologically relevant change), with indistinguishable %G+C profiles. Wheat was grown to maturity (129 d) in a different soil microcosm design, and rhizosphere microbial communities from plants grown at 350 and 700 μl CO2 l−1 were identical according to the DNA analyses. In these experiments rhizosphere microbial community structure at the broad scale was unaffected by the interactions occurring below ground as a result of elevated concentrations of CO2.