Microbial immobilisation and turnover of C labelled substrates in two arable soils under field and laboratory conditions

Microbial biomass C immobilisation and turnover were studied under field and laboratory conditions in soils of high yield (HY) and low yield (LY) areas within an agricultural field. We compared the size and activity of soil microbial biomass (SMB) in the soils of the different yield areas under field and laboratory conditions. Soils were amended with ¹³C labelled mustard (Sinapis alba) residues (both experiments) and labelled glucose (laboratory only) at 500 μg C g⁻¹ dry soil. SMB-C, dissolved organic carbon (DOC) and total C content were monitored in the field and the laboratory. CO₂-efflux was also measured in laboratory treatments. Isotope ratios were determined for SMB in both experiments, but other variables only in the laboratory treatments. A positive priming effect was measured in three of four laboratory treatments. Priming was induced after a significant increase of soil derived C in the microbial biomass. Thereafter, the total C loss through priming was always smaller than or equal to the decline in microbial biomass C. In field and laboratory experiments SMB in the HY soil immobilised less of the added substrate C than LY soil SMB. Calculated turnover times in the laboratory glucose amendment were 0.24 (HY) and 0.31 y (LY), in the laboratory mustard treatment 0.58 (HY) and 0.44 y (LY) and in the field mustard amendments 1.09 (HY) and 1.25 y (LY). In both the field mustard and laboratory glucose treatments turnover in the HY soil tended to exceed that in the LY soil. These turnover times as well as the reaction of SMB-C to drying-rewetting and substrate addition, indicated that the HY soil possessed a more active microbial community with a more rapid C turnover than the LY soil. As C turnover is considered to be closely linked to nutrient cycles, faster turnover in the HY soil may involve a better nutrient supply for crops resulting in higher agricultural yield.     

Extractability of labelled microbial biomass N by electro-ultrafiltration and CaCl2 extraction

A laboratory soil incubation and a pot experiment with ryegrass were carried out in order to examine the extractability of microbial biomass N by using either 10-mM CaCl₂ extraction or the electro-ultrafiltration (EUF) method. The aim of the experiment was to test the hypothesis whether the organic N (Norg) extracted by EUF or CaCl₂ from dried soil samples represents a part of the microbial biomass. For the laboratory incubation a ¹⁵N-labelled Escherichia coli suspension was mixed with the soil. For the pot experiment a suspension of ¹⁵N-labelled bacteria was applied which had previously been isolated from the soil used. Soil samples of both treatments, with and without applied bacterial suspension, were extracted by EUF and CaCl₂. The extractability of applied microbial biomass was estimated from the difference in extractable Norg between the two treatments. In addition, the N isotopic composition in the upper plant matter, in the soil, and in organic and inorganic N fractions of EUF and CaCl₂ extracts was analysed. Both experiments showed that the applied microbial biomass was highly accessible to mineralization and thus represented potentially mineralizable N. However, this mineralizable N was not extractable by CaCl₂ or by the EUF method. It was, therefore, concluded that the organic N released on soil drying and which was thus extractable was derived from the non-biomass soil organic matter. The result suggests that both extraction methods may provide a suitable index for mineralizable N only in cases where the decomposable organic substrates are derived mainly from sources other than the living soil biota.Dedicated to Professor J. C. G. Ottow on the occasion of his 60th birthday     

Ammonium immobilisation during chloroform fumigation

The assumption in using the chloroform fumigation technique for microbial biomass determination is that microbes are killed or at least inactivated by the treatment. Problems associated with transformations of the N released on or during fumigation have so far only been associated with the fumigation-incubation method. A laboratory and a field study were carried out to investigate the possible N transformations during biomass determination by the fumigation-extraction method. Labelled NH₄NO₃ (either the NO₃⁻, NH₄⁺ or both pools were ¹⁵N enriched) was applied to the soil and biomass determinations made at intervals subsequently. The size and enrichment of the ammonium (NH₄⁺), and nitrate (NO₃⁻) pools were determined before and after chloroform fumigation. The ¹⁵N enrichment of the NH₄⁺ pool after fumigation could only be explained if immobilisation of ammonium occurred at some time during the 24 h fumigation period. The extent of this immobilisation was calculated. In addition, there was evidence that nitrification occurred during the fumigation procedure at the start of the laboratory study and throughout the field study. The laboratory and field study differed mainly in the dynamics related to NO₃⁻ uptake and release. There was evidence for uptake of NO₃⁻ by the microbial biomass with and without utilization. We conclude that the ¹⁵N enrichment in the microbial biomass cannot be accurately determined when N transformations and release of non-utilized N occurs during fumigation. The possible immobilisation of mineral N during fumigation will affect the magnitude of the factor used to convert measured microbial biomass N to actual microbial biomass N in soil.     

Gross N transfer rates in field soils measured by 15N-pool dilution

Abstract

A micro-plot ¹⁵N-tracer experiment was established in three different soils of a long-term soil fertility field experiment. The nutrient-poor loam sand has been subjected to various treatments over the years and this has resulted in different organic C (0.35% – 0.86%), microbial biomass (38.3 – 100.0 µg C _mic g^?1 soil), clay and fine silt contents. Using the ¹⁵N-pool dilution technique, we assessed gross N-transfer rates in the field. Gross N mineralization rates varied strongly among the three plots and ranged between 0.4 and 4.2 µg N g^?1 soil d^?1. Gross nitrification rates were estimated to be between 0 and 2.1 µg N g^?1 soil d^?1. No correlation between gross N mineralization rates and the organic matter content of the soils was established. However, gross nitrate consumption rates increased with increasing soil C content. The ¹⁵N-pool dilution technique was successfully used to measure gross N transfer rates directly in the field.     

Microbial immobilization of ammonium and nitrate in relation to ammonification and nitrification rates in organic and conventional cropping systems

Agricultural systems that receive high or low organic matter (OM) inputs would be expected to differ in soil nitrogen (N) transformation rates and fates of ammonium (NH₄⁺) and nitrate (NO₃⁻). To compare NH₄⁺ availability, competition between nitrifiers and heterotrophic microorganisms for NH₄⁺, and microbial NO₃⁻ assimilation in an organic vs. a conventional irrigated cropping system in the California Central Valley, chemical and biological soil assays, ¹⁵N isotope pool dilution and ¹⁵N tracer techniques were used. Potentially mineralizable N (PMN) and hot minus cold KCl-extracted NH₄⁺ as indicators of soil N supplying capacity were measured five times during the tomato growing season. At mid-season, rates of gross ammonification and gross nitrification after rewetting dry soil were measured in microcosms. Microbial immobilization of NO₃⁻ and NH₄⁺ was estimated based on the uptake of ¹⁵N and gross consumption rates. Gross ammonification, PMN, and hot minus cold KCl-extracted NH₄⁺ were approximately twice as high in the organically than the conventionally managed soil. Net estimated microbial NO₃⁻ assimilation rates were between 32 and 35% of gross nitrification rates in the conventional and between 37 and 46% in the organic system. In both soils, microbes assimilated more NO₃⁻ than NH₄⁺. Heterotrophic microbes assimilated less NH₄⁺ than NO₃⁻ probably because NH₄⁺ concentrations were low and competition by nitrifiers was apparently strong. The high OM input organic system released NH₄⁺ in a gradual manner and, compared to the low OM input conventional system, supported a more active microbial biomass with greater N demand that was met mainly by NO₃⁻ immobilization.     

Turnover of grain legume N rhizodeposits and effect of rhizodeposition on the turnover of crop residues

The turnover of N derived from rhizodeposition of faba bean (Vicia faba L.), pea (Pisum sativum L.) and white lupin (Lupinus albus L.) and the effects of the rhizodeposition on the subsequent C and N turnover of its crop residues were investigated in an incubation experiment (168 days, 15 °C). A sandy loam soil for the experiment was either stored at 6 °C or planted with the respective grain legume in pots. Legumes were in situ ¹⁵N stem labelled during growth and visible roots were removed at maturity. The remaining plant-derived N in soil was defined as N rhizodeposition. In the experiment the turnover of C and N was compared in soils with and without previous growth of three legumes and with and without incorporation of crop residues. After 168 days, 21% (lupin), 26% (faba bean) and 27% (pea) of rhizodeposition N was mineralised in the treatments without crop residues. A smaller amount of 15–17% was present as microbial biomass and between 30 and 55% of mineralised rhizodeposition N was present as microbial residue pool, which consists of microbial exoenzymes, mucous substances and dead microbial biomass. The effect of rhizodeposition on the C and N turnover of crop residues was inconsistent. Rhizodeposition increased the crop residue C mineralisation only in the lupin treatment; a similar pattern was found for microbial C, whereas the microbial N was increased by rhizodeposition in all treatments. The recovery of residual ¹⁵N in the microbial and mineral N pool was similar between the treatments containing only labelled crop residues and labelled crop residues + labelled rhizodeposits. This indicates a similar decomposability of both rhizodeposition N and crop residue N and may be attributable to an immobilisation of both N sources (rhizodeposits and crop residues) as microbial residues and a subsequent remineralisation mainly from this pool.Abbreviations C or N_dec C or N decomposed from residues - C or N_mic microbial C or N - C or N_micres microbial residue C or N - C or N_min mineralised C or N - C or N_input added C or N as crop residues and/or rhizodeposits - dfr derived from residues - dfR derived from rhizodeposition - Ndfr N derived from residues - NdfR N derived from rhizodeposition - N_loss losses of N derived from residues - SOM soil organic matter - WHC water holding capacity     

C and N turnover in structurally intact soils of different texture

The turnover of native and applied C and N in undisturbed soil samples of different texture but similar mineralogical composition, origin and cropping history was evaluated at −10 kPa water potential. Cores of structurally intact soil with 108, 224 and 337 g clay kg⁻¹ were horizontially sliced and ¹⁵N-labelled sheep faeces was placed between the two halves of the intact core. The cores together with unamended treatments were incubated in the dark at 20 °C and the evolution of CO₂-C determined continuously for 177 d. Inorganic and microbial biomass N and ¹⁵N were determined periodically. Net nitrification was less in soil amended with faeces compared with unamended soil. When adjusted for the NO₃-N present in soil before faeces was applied, net nitrification became negative indicating that NO₃-N had been immobilized or denitrified. The soil most rich in clay nitrified least N and ¹⁵N. The amounts of N retained in the microbial biomass in unamended soils increased with clay content. A maximum of 13% of the faeces ¹⁵N was recovered in the microbial biomass in the amended soils. CO₂-C evolution increased with clay content in amended and unamended soils. CO₂-C evolution from the most sandy soil was reduced due to a low content of potentially mineralizable native soil C whereas the rate constant of C mineralization rate peaked in this soil. When the pool of potentially mineralizable native soil C was assumed proportional to volumetric water content, the three soils contained similar proportions of potentially mineralizable native soil C but the rate constant of C mineralization remained highest in the soil with least clay. Thus although a similar availability of water in the three soils was ensured by their identical matric potential, the actual volume of water seemed to determine the proportion of total C that was potentially mineralizable. The proportion of mineralizable C in the faeces was similar in the three soils (70% of total C), again with a higher rate constant of C mineralization in the soil with least clay. It is hypothesized that the pool of potentially mineralizable C and C rate constants fluctuate with the soil water content.     

Priming effect of biuret addition on native soil N mineralisation under laboratory conditions

This study was carried out to quantify the priming effect of biuret on native soil nitrogen (N) mineralisation during a 112-day incubation. Addition of biuret (100 mg ¹⁵N-labelled biuret kg⁻¹ soil) increased the turnover rate constant of soil organic matter and had a positive priming effect on native soil N mineralisation in two soils. The additional mineralisation was 0.65% of the total soil N (equivalent to 47.1 kg N ha⁻¹) in a sandy loam soil and 0.62% of the soil N (equivalent to 46.5 kg N ha⁻¹) in a silt loam soil.     

Cold storage and laboratory incubation of intact soil cores do not reflect in-situ nitrogen cycling rates of tropical forest soils

Measurements of N transformation rates in tropical forest soils are commonly conducted in the laboratory from disturbed or intact soil cores. On four sites with Andisol soils under old-growth forests of Panama and Ecuador, we compared N transformation rates measured from laboratory incubation (at soil temperatures of the sites) of intact soil cores after a period of cold storage (at 5 °C) with measurements conducted in situ. Laboratory measurements from stored soil cores showed lower gross N mineralization and NH₄⁺ consumption rates and higher gross nitrification and NO₃⁻ immobilization rates than the in-situ measurements. We conclude that cold storage and laboratory incubation change the soils to such an extent that N cycling rates do not reflect field conditions. The only reliable way to measure N transformation rates of tropical forest soils is in-situ incubation and mineral N extraction in the field.     

Gross sulphur mineralisation-immobilisation turnover in soil amended with plant residues

The rates of sulphur (S) released to and removed from the soil inorganic pool were estimated using the isotopic dilution technique. In an initial study fresh soil was mixed with combinations of two inorganic S levels (0 and 10 μg S g⁻¹ soil) and three plant residues (wheat straw, perennial ryegrass and oilseed rape) and followed over 32 days of incubation. As ³⁵S recovery was inadequate prior to day 2 and re-mineralisation of immobilised ³⁵S occurred after day 8 thereby invalidating the method, estimates of gross S transformation rates should be based on data sampled between days 2 and 8. In the main experiment 16 plant residues with ranges in S contents of 0.08-0.81%, C/S ratios of 50-604 and lignin content of 0.9-10.8 were mixed with soil and carrier-free ³⁵S label. Net turnover rates varied from 58% of S in Persian clover being immobilised to 76% of S in winter cress being mineralised within 5 days of incubation. Gross S mineralisation varied from 0.9-14.9 μg S g⁻¹ soil d⁻¹, whereas gross immobilisation only varied from 0.5 to 3.1 μg S g⁻¹ d⁻¹. Gross S immobilisation was strongly correlated to the C/S ratio of the plant material (P<0.001), whereas gross S mineralisation showed a weaker, but still significant, correlation with lignin content (P<0.05). The results indicate that immobilisation may predominantly have been a biological process in response to carbon addition while early mineralisation may have been dominated by the biochemical hydrolysis of organic sulphates in the residues. If attention is paid to the various constraints and limitations, isotopic pool dilution using ³⁵S offers a tool that may prove valuable in understanding and modelling soil S turnover.     

15N isotope dilution methodology for evaluating the dynamics of biologically fixed N in legume-non-legume associations

We examined the theoretical basis for estimating the transfer of N₂ fixed by legumes to companion cereals or grasses in intercropping or pasture systems using ¹⁵N isotope dilution methodology. A method was developed to calculate the symbiotic dependence of the legume in a mixed stand based on ¹⁵N enrichment of the associated non-legume and the estimate of fixed N transfer. Published field data were used to illustrate the application of the method. Complementary treatments for verifying N transfer and options for increasing the accuracy of estimates of N transfer are discussed.     

好氧条件下两种水稻土的氮素总转化速率比较

Although to date individual gross N transformations could be quantified by ~(15)N tracing method and models,studies are still limited in paddy soil.An incubation experiment was conducted using topsoil(0-20 cm) and subsoil(20-60 cm) of two paddy soils,alkaline and clay(AC) soil and neutral and silt loam(NSL) soil,to investigate gross N transformation rates.Soil samples were labeled with either ~(15)NH4_NO_3 or NH_4~(15)NO_3,and then incubated at 25 °C for 168 h at 60%water-holding capacity.The gross N mineralization(recalcitrant and labile organic N mineralization) rates in AC soil were 1.6 to 3.3 times higher than that in NSL soil,and the gross N nitrification(autotrophic and heterotrophic nitrification) rates in AC soil were 2.4 to 4.4 times higher than those in NSL soil.Although gross NO_3~- consumption(i.e.,NO_3~- immobilization and dissimilatory NO_3~- reduction to NH_4~+ rates increased with increasing gross nitrification rates,the measured net nitrification rate in AC soil was approximately 2.0 to 5.1 times higher than that in NSL soil.These showed that high NO_3~- production capacity of alkaline paddy soil should be a cause for concern because an accumulation of NO_3~- can increase the risk of NO_3~- loss through leaching and denitrification.     

Gross N mineralization rates after application of composted grape marc to soil

Grape marc is a common waste product of the wine production industry. When partially composted and applied to soil it may contain enough N to affect vine growth and hence wine quality. Yet little is known about the quantity and timing of N release from composted grape marc. A laboratory incubation was conducted where composted grape marc amended and non-amended soils were periodically sampled over 148 days at 15 °C for gross N mineralization rates, C mineralization and microbial biomass-C. Gross N mineralization rates were determined by ¹⁵N pool dilution using both analytical equations and the numerical model FLUAZ (Mary, B., Recous, S., Robin, D., 1998. A model for calculating nitrogen fluxes in soil using 15N tracing. Soil Biology & Biochemistry 30, 1963-1979.). Both analytical and FLUAZ determined gross N mineralization rates were in close agreement in the control soil. However, in composted grape marc amended soils there was a discrepancy between the two solutions. Findings indicate that composted grape marc caused a net immobilization of N for the first 50-days of incubation, after which enough N was released to require consideration in fertilizer-N strategies.     

Effect of chemical composition on the release of nitrogen from agricultural plant materials decomposing in soil under field conditions

Summary In two field experiments, plant materials labelled with ¹⁵N were buried separately within mesh bags in soil, which was subsequently sown with barley. In the first experiment, different parts of white clover (Trifolium repens), red clover (T. pratense), subterranean clover (T. subterraneum), field bean (Vicia faba), and timothy (Phleum pratense) were used, and in the second, parts of subterranean clover of different maturity. The plant materials were analysed for their initial concentrations of total N, ¹⁵N, C, ethanol-soluble compounds, starch, hemicellulose, cellulose, lignin, and ash. After the barley had been harvested, the bags were collected and analysed for their total N and ¹⁵N. In the first experiment the release of N was highest from white clover stems + petioles (86%) and lowest from field bean roots (20%). In stepwise regression analysis, the release of N was explained best by the initial concentrations of lignin, cellulose, hemicellulose, and N (listed according to decreasing partial correlations). Although the C/N ratio of the plant materials varied widely (11–46), statistically the release of N was not significantly correlated with this variable. The results of the second experiment using subterranean clover of different maturity confirmed those of the first experiment.     

Mobilization of labelled N in soybean leaves during early podfill

Soybeans accumulate N in vegetative tissues up to pod initiation after which total vegetative N may remain constant during early phases of pod development. Eventually much of the vegetative N is mobilized to the pods. The mobilization of N from vegetative tissue to pods during the first few days of pod development is poorly understood but is important to an overall understanding of soybean N nutrition. The vegetative tissues of field grown soybeans were labelled with ¹⁵N and sampled weekly during the reproductive phase of plant growth. Three foliar applications of (¹⁵NH₄)₂SO₄ were made prior to pod initiation at a combined rate of 3.3 kg N/ha. To immobilize soil N and to increase soybean dependance on N₂‐fixation, sawdust was applied at a rate of 52 t ha^‐1 . Irrigation was required almost weekly because of a shallow soil profile and below normal summer precipitation. Mobilization of vegetative N began immediately upon pod initiation and continued at a linear rate through pod development. It appeared that N₂‐fixation was able to provide approximately half of the N in pods during early podfill. Nitrogen content of vegetative tissue declined as soon as pods began developing.     

Microbially-mediated gross N transfer rates in field soils in relation to physico-chemical transfer processes

Abstract

Measurements of gross N transfer in soils have as yet not distinguished between biological or physico-chemical processes. Here, we present a new approach that allows microbially-mediated gross N transfer rates to be estimated in undisturbed soils without adding ¹⁵N. It is based on the assumption that in undisturbed soil, the soil microbial growth rate is equal to its death rate. To assess the contribution of biological versus physico-chemical N transfer processes, we combined the new approach with the ¹⁵N-pool dilution technique. The relationship between both processes varied with soil C and fine particle contents. Nearly equal rates were observed within the carbon-poor soil (0.35% C_org, low fine particle content), whereas up to 2.5 times higher physico-chemical than biological N transfer rates were measured within the carbon-enriched soil (0.86% C_org, higher fine particle content). Furthermore, microbially-mediated gross N transfer rates increased three-fold after N fertilization compared to the unfertilized control.     

Microbial biomass and organic matter turnover in wetland rice soils

A decline in rice yields has been associated with intensification of rice production. In continuously irrigated systems this has been attributed to a decline in soil N supply. Nutrient mineralisation and immobilisation is constrained by the quantity and nature of the organic substrates and the physico-chemical environment of the soil system itself. A flooded soil is very different from an aerobic one; electron acceptors other than oxygen have to be used. The transition to continuously anaerobic conditions associated with the intensification of wetland rice systems affects their organic matter turnover and may adversely affect their productivity.     

Comparison of NH4 pool dilution techniques to measure gross N fluxes in a coarse textured soil

We compared gross N fluxes by ¹⁵N pool dilution in a coarse-textured agricultural soil when ¹⁵N was applied to the soil NH₄⁺ pool by either: (i) mixing a ¹⁵NH₄NO₃ solution into disturbed soil or (ii) injection of ¹⁵NH₃ gas into intact soil cores. The two techniques produced similar results for gross N mineralization rates indicating that NH₄⁺ production in soil was not altered by soil disturbance, method of application (gas vs. solution), or amount of N applied. This was not the case for immobilization rates, which were twofold higher when ¹⁵N label was applied to the soil NH₄⁺ pool with the mixing technique compared to the injection technique. This was attributed to the fact that more NH₄⁺ was applied with the mixing technique. Estimates of gross nitrification were accompanied by large error terms meaning differences between ¹⁵N labeling methods could not be accurately assessed for this process rate.     

Rhizodeposition of C and N in peas and oats after C-N double labelling under field conditions

Compounds released by plant roots during growth can make up a high proportion of below-ground plant (BGP) carbon and nitrogen, and therefore influence soil organic matter turnover and plant nutrient availability by stimulating the soil microorganisms. The present study was conducted to examine the amount and fate of C (CdfR) and N rhizodeposits (NdfR), in this study defined as root-derived C or N present in the soil after removal of roots and root fragments, released during reproductive growth. BGP biomass of peas (Pisum sativum L.) and oats (Avena sativa L.) was successfully labelled in situ with a ¹³C-glucose-¹⁵N-urea mixture under field conditions using a stem feeding method. Pea plants were labelled at the beginning of flowering and harvested 36 and 52 days after labelling at pod filling (P_P) and maturity (P_M), respectively. Oat plants were labelled at grain filling and harvested 42 days after labelling at maturity (O_M). CdfR was 24.2% (P_P), 29.6% (P_M) and 30.8% (O_M) of total recovered plant C. NdfR was 32.1% (P_P), 36.4% (P_M) and 30.0% (O_M) of total plant N. Due to higher N assimilation, amounts of NdfR were four times higher in peas in comparison with oats. The results for NdfR in peas were higher than results from other studies. The C-to-N ratio of rhizodeposits was lower under peas (17.3) than under oats (41.9) at maturity. At maturity, microbial CdfR at 0-30 cm soil depth was 37% of the microbial biomass C in peas and 59% in oats. Microbial NdfR was 15% of microbial N in peas and 5% in oats. Furthermore, inorganic NdfR was 34% in peas and 9% in oats at 0-30 cm at maturity. These results show that rhizodeposits of peas provide a more easily available substrate to soil microorganisms, which are incorporated to a greater extent and turned over faster in comparison with oats. Beside the higher amounts of N released from pea roots, this process contributes to the higher N-availability for subsequent crops.     

Microbial immobilisation of C rhizodeposits in rhizosphere and root-free soil under continuous C labelling of oats

A greenhouse experiment was conducted by growing oats (Avenasativa L.) in a continuously ¹³CO₂ labeled atmosphere. The allocation of ¹³C-labeled photosynthates in plants, microbial biomass in rhizosphere and root-free soil, pools of soil organic C, and CO₂ emissions were examined over the plant's life cycle. To isolate rhizosphere from root-free soil, plant seedlings were placed into bags made of nylon monofilament screen tissue (16 μm mesh) filled with soil. Two peaks of ¹³C in rhizosphere pools of microbial biomass and dissolved organic carbon (DOC), as well as in CO₂ emissions at the earing and ripeness stages were revealed. These ¹³C maxima corresponded to: (i) the end of rapid root growth and (ii) beginning of root decomposition, respectively. The δ¹³C values of microbial biomass were higher than those of DOC and of soil organic matter (SOM). The microbial biomass C accounted for up to 56 and 39% of ¹³C recovered in the rhizosphere and root-free soil, respectively. Between 4 and 28% of ¹³C assimilated was recovered in the root-free soil. Depending on the phenological stage, the contribution of root-derived C to total CO₂ emission from soil varied from 61 to 92% of total CO₂ evolved, including 4-23% attributed to rhizomicrobial respiration. While 81-91% of C substrates used for microbial growth in the root-free soil and rhizosphere came from SOM, the remaining 9-19% of C substrates utilized by the microbial biomass was attributable to rhizodeposition. The use of continuous isotopic labelling and physical separation of root-free and rhizosphere soil, combined with natural ¹³C abundance were effective in gaining new insight on soil and rhizosphere C-cycling.