Time-dependent sorption of imidacloprid in two different soils

Time-dependent sorption of imidacloprid [1-[(6-chloro-3-pyridinyl)-methyl]-N-nitro-2-imidazolidinimine] was investigated with two German soils (sandy loam and silt loam). Soil batches containing the active ingredient (0.33 mg/kg) were incubated for 100 days. After selected aging periods, imidacloprid desorbed by 0.01 M CaCl(2) (soluble phase) and by organic solvents (methanol and acetonitrile) and reflux extraction with acidified methanol (sorbed phase) was determined. Calculated sorption coefficients K(d) and K(oc) increased by a factor of 3.2-3.8 during 100 days of aging. Additionally, the time-dependent sorption was verified by a column leaching experiment with the aged soil. The amount of imidacloprid in column eluates (0.01 M CaCl(2)) decreased compared to total recovered by a factor of approximately 2. Sorption of imidacloprid thus increased with residence time in soil, making it more resistant to leaching. These results are further information to explain the low leaching potential of imidacloprid in the field, despite its high water solubility.     

Nitrogen losses from two grassland soils with different fungal biomass

Nitrogen losses from agricultural grasslands cause eutrophication of ground- and surface water and contribute to global warming and atmospheric pollution. It is widely assumed that soils with a higher fungal biomass have lower N losses, but this relationship has never been experimentally confirmed. With the increased interest in soil-based ecosystem services and sustainable management of soils, such a relationship would be relevant for agricultural management. Here we present a first attempt to test this relationship experimentally. We used intact soil columns from two plots from a field experiment that had consistent differences in fungal biomass (68 ± 8 vs. 111 ± 9 μg C g⁻¹) as a result of different fertilizer history (80 vs. 40 kg N ha⁻¹ y⁻¹ as farm yard manure), while other soil properties were very similar. We performed two greenhouse experiments: in the main experiment the columns received either mineral fertilizer N or no N (control). We measured N leaching, N₂O emission and denitrification from the columns during 4 weeks, after which we analyzed fungal and bacterial biomass and soil N pools. In the additional ¹⁵N experiment we traced added N in leachates, soil, plants and microbial biomass. We found that in the main experiment, N₂O emission and denitrification were lower in the high fungal biomass soil, irrespective of the addition of fertilizer N. Higher ¹⁵N recovery in the high fungal biomass soil also indicated lower N losses through dentrification. In the main experiment, N leaching after fertilizer addition showed a 3-fold increase compared to the control in low fungal biomass soil (11.9 ± 1.0 and 3.9 ± 1.0 kg N ha⁻¹, respectively), but did not increase in high fungal biomass soil (6.4 ± 0.9 after N addition vs. 4.5 ± 0.8 kg N ha⁻¹ in the control). Thus, in the high fungal biomass soil more N was immobilized. However, the ¹⁵N experiment did not confirm these results; N leaching was higher in high fungal biomass soil, even though this soil showed higher immobilization of ¹⁵N into microbial biomass. However, only 3% of total ¹⁵N was found in the microbial biomass 2 weeks after the mineral fertilization. Most of the recovered ¹⁵N was found in plants (approximately 25%) and soil organic matter (approximately 15%), and these amounts did not differ between the high and the low fungal biomass soil. Our main experiment confirmed the assumption of lower N losses in a soil with higher fungal biomass. The additional ¹⁵N experiment showed that higher fungal biomass is probably not the direct cause of higher N retention, but rather the result of low nitrogen availability. Both experiments confirmed that higher fungal biomass can be considered as an indicator of higher nitrogen retention in soils.     

Carbon-nitrogen relationships during the humification of cellulose in soils containing different amounts of clay

¹⁴C-labelled cellulose and ¹⁵N-labelled (NH₄)₂SO₄ were added to four soils with clay contents of 4, 11, 18 and 34%, respectively. Labelled cellulose was added to each soil in amounts corresponding to 1, 2 and 4 mg C g^?1 soil, respectively, and labelled NH₄⁺ at the rate of 1 mg N per 25 mg labelled C.After the first month of incubation at temperatures of 10, 20 and 30°C, respectively, from 38 to 65% of the labelled C added in cellulose had disappeared from the soils as CO₂, and from 60 to nearly 100% of the labelled N added as NH₄⁺ were incorporated into organic forms. The ratio of labelled C remaining in the soils to labelled N in organic forms was close to 25 after 10 days of incubation, decreasing to about 15 after 1 month and about 10 after 4 yr.The retention of total labelled C was largest in the soil with the highest content of clay where after 4 yr it was 25% of that added, compared to 12 in the soil with the lowest content of clay. The incorporation of labelled N in organic forms and its retention in these forms was not directly related to the content of clay in the soils, presumably because the two soils with the high content of clay had a relatively high content of available unlabelled soil-N which was used for synthesis of metabolic material.The proportionate retention of labelled C for a given soil was largely independent of the size of the amendments, whereas the proportionate amount of labelled N incorporated into organic forms increased in the clay-rich soils with increasing size of amendments. Presumably this is because the dilution with unlabelled soil-N was less with the large amendments.From 50 to 70% of the total labelled C remaining in the soils after the first month of incubation was acid hydrolyzable, as compared to 80–100% of the total remaining labelled organic N. This relationship held throughout the incubation and was independent of the size of the amendment and of the temperature of incubation.During the second, third and fourth year of incubation the half-life of labelled amino acid-N in the soils was longer than the half-life of labelled amino acid-C, presumably due to immobilization reactions. Some of the labelled organic N when mineralized was re-incorporated into organic compounds containing increasing proportions of native soil-C. whereas labelled C when mineralized as CO₂ disappeared from the soils.In general, native C and native organic N were less acid hydrolyzable and were accounted for less in amino acid form than labelled C and N.The amount of labelled amino acid-C, formed during decomposition of the labelled cellulose, and retained in the soil, was proportional to the clay content. This amount was about three times as large in the soil with the highest content of clay as in the soil with the lowest content. This difference between the soils was established during the first 10 days of incubation when biological activity was most intense, and it held throughout the 4 yr of incubation; proportionally it was independent of the amount of cellulose added and the temperature.In contrast, the labelled amino acid-N content was not directly related to the amount of clay in the soil, presumably because more unlabelled soil-N was available for synthesis of metabolic material in the two clay-rich soils than in those soils with less clay. The wider ratio between labelled amino acid-C and labelled amino acid-N in the two clay-rich soils as compared with those obtained with the soils with less clay indicates this.The effect of clay in increasing the content of organic matter in soil is possibly caused by newly synthesized matter, extracellular metabolites, as well as cellular material, forming biostable complexes and aggregates with clay. The higher the concentration of clay the more readily the interactions take place. The presence of clay may also increase the efficiency of using substrate for synthesis.     

Fluctuations in microbial biomass indices at different sampling times in soils from tussock grasslands

Microbial biomass in four topsoils from New Zealand tussock grasslands was estimated by three biochemical procedures at five sampling times over a 15 month period. In Conroy, Cluden and Tima soils, biomass C content was high in two sets of March (summer-autumn) samples and low in October (early spring) samples; in Carrick soil from a wetter, cooler environment, it was similar at all sampling times. Significant time-of-sampling variations occurred with Min-N flush in Tima and Carrick soils, and with adenosine 5'-triphosphate (ATP) content in three of the soils. Generally, the ratios of these biomass indices also varied significantly at some sampling times. Because of this variability, common factors could not validly be used with these soils for estimating biomass C contents from Min-flush or ATP values.The contribution of bacteria and fungi to the respiratory activity of the microbial biomass was unsuccessfully investigated using streptomycin and actidione as differential inhibitors of anabolic metabolism in the presence of added glucose. In three of the soils, rates of O₂ uptake did not generally increase significantly during incubation, even with added N, P, K and S or prior incubation overnight. In Conroy soil, rates did increase significantly, but the effects of the antibiotics separately and together could not be satisfactorily balanced.     

Turnover time of microbial biomass carbon in Japanese upland soils with different textures

We investigated the turnover time of microbial biomass-C in Japanese upland soils with various textures and examined the soil physicochemical properties influencing their turnover time. Samples from five different soil types (sand-dune regosol, light-colored Andosol, humic Andosol, brown forest soil, and dark red soil) were taken from upland concrete-frame plots in the experimental field of Chiba University. Each soil amended with [U -¹³C] glucose was incubated for 80 d at 25°C. Microbial biomass-C and -¹³C in soil were periodically determined by the fumigation-extraction method. The longest turnover time of microbial biomass-C was observed in the dark red soil (215 d) followed by the humic Andosol (134 d), brown forest soil (97 d), and light-colored Andosol (83 d) and the shortest in the sand-dune regosol (45 d). The turnover time of microbial biomass-C was significantly correlated with the value of soil clay (R: 0.917*), CEC (R: 0.921*), and macroaggregate (R: 0.907*) contents, but not with the total-C content. The amount of microbial biomass-C showed a close correlation with the turnover time of microbial biomass-C, suggesting that the turnover time of microbial biomass-C is an important factor influencing the accumulation of microbial biomass-C in soil.     

Different sorption behaviors for wine polyphenols in contact with oak wood

The evolution of polyphenols of enological interest- monomeric anthocyanins, (+)-catechin, (-)-epicatechin, gallic acid, and trans-resveratrol-in the presence of oak wood was investigated in aging-model conditions. Disappearance kinetics showed that, except for gallic acid, all of the wine polyphenols tend to disappear from the model wine in presence of oak wood, to reach an equilibrium after 20 days of contact. At equilibrium, the higher disappearance rates were obtained for monomeric anthocyanins and trans-resveratrol with values of 20 and 50%, respectively. For monomeric anthocyanins, the rate of disappearance seemed to be independent of their nature. In order to evaluate the contribution of sorption to oak wood in the disappearance phenomena, sorption kinetics were determined for trans-resveratrol and malvidin-3-glucoside through the extraction and the quantification of the fraction sorbed to wood. These curves showed that the wood intake of trans-resveratrol and malvidin-3-glucoside followed a two-step behavior, with a higher rate during the first 2 days, likely due to a surface sorption mechanism, and then a slower rate to reach the equilibrium, which could be related to a diffusion mechanism. The comparison of disappeared and sorbed amounts at equilibrium showed that a minor part of the disappeared monomeric anthocyanins were sorbed by wood. In contrast, half of the concentration decrease of trans-resveratrol in wine finds its origin in a sorption mechanism by oak wood. Results in real wine show similar sorption kinetics.     

Phosphorus sorption capacities and saturation of soils in two regions with different livestock densities in northwest Germany

Abstract. Soils in areas with high livestock density contribute to the eutrophication of aquatic ecosystems through loss of nutrients, especially phosphorus (P). In order to identify the potential for P loss from such soils we determined phosphorus extracted by water (H₂O-P), by double lactate (DL-P), and P sorption capacity (PSC) and degree of P saturation (DPS) in soil samples from two counties, one with low (Harle-catchment) and the other with very high livestock density (Vechta). Both catchments are hydrologically connected with the tidal areas of the North Sea.
The mean concentrations of H₂O-P (0.4mmol/kg) and DL-P (3.9 mmol/kg) were lower in the Harle-catchment than in the Vechta area (1.2 mmol/kg, 6.8mmol/kg). Although oxalate-extractable Al (Al_ox) and Fe (Fe_ox) and the derived PSCs varied according to soil type and to land use, the livestock density and the resulting high concentrations of oxalate-extractable P (P_ox) were shown to be the main reason for the very high DPS of up to 179% in the county of Vechta. These values exceeded DPS reported from other intensive pig feeding areas in western Europe and indicate the potential for significant P loss. Less than 40% of the variation in P_ox could be explained by the routinely determined H₂O-Por DL-P. Geostatistical analyses indicated that the spatial variability of P_ox depended on manurial history of fields and Al_ox, showed still smaller-scale variability. These were the major constraints for regional assessments of P losses and eutrophication risk from agricultural soils using available soil P-test values, digital maps and geostatistical methods.     

Gross nitrogen mineralization and nitrification rates and their relationships to enzyme activities and the soil microbial biomass in soils treated with dairy shed effluent and ammonium fertilizer at different water potentials

 Gross N mineralization and nitrification rates and their relationships to microbial biomass C and N and enzyme (protease, deaminase and urease) activities were determined in soils treated with dairy shed effluent (DSE) or NH₄ ⁺ fertilizer (NH₄Cl) at a rate equivalent to 200 kg N ha^–1 at three water potentials (0, –10 and –80 kPa) at 20  °C using a closed incubation technique. After 8, 16, 30, 45, 60 and 90 days of incubation, sub-samples of soil were removed to determine gross N mineralization and nitrification rates, enzyme activities, microbial biomass C and N, and NH₄ ⁺ and NO₃ ^– concentrations. The addition of DSE to the soil resulted in significantly higher gross N mineralization rates (7.0–1.7 μg N g^–1 soil day^–1) than in the control (3.8–1.2 μg N g^–1 soil day^–1), particularly during the first 16 days of incubation. This increase in gross mineralization rate occurred because of the presence of readily mineralizable organic substrates with low C : N ratios, and stimulated soil microbial and enzymatic activities by the organic C and nutrients in the DSE. The addition of NH₄Cl did not increase the gross N mineralization rate, probably because of the lack of readily available organic C and/or a possible adverse effect of the high NH₄ ⁺ concentration on microbial activity. However, nitrification rates were highest in the NH₄Cl-treated soil, followed by DSE-treated soil and then the control. Soil microbial biomass, protease, deaminase and urease activities were significantly increased immediately after the addition of DSE and then declined gradually with time. The increased soil microbial biomass was probably due to the increased available C substrate and nutrients stimulating soil microbial growth, and this in turn resulted in higher enzyme activities. NH₄Cl had a minimal impact on the soil microbial biomass and enzyme activities, possibly because of the lack of readily available C substrates. The optimum soil water potential for gross N mineralization and nitrification rates, microbial and enzyme activities was –10 kPa compared with –80 kPa and 0 kPa. Gross N mineralization rates were positively correlated with soil microbial biomass N and protease and urease activities in the DSE-treated soil, but no such correlations were found in the NH₄Cl-treated soil. The enzyme activities were also positively correlated with each other and with soil microbial biomass C and N. The forms of N and the different water potentials had a significant effect on the correlation coefficients. Stepwise regression analysis showed that protease was the variable that most frequently accounted for the variations of gross N mineralization rate when included in the equation, and has the potential to be used as one of the predictors for N mineralization. Received: 10 March 1998     

Feasibility of a cell separation-proteomic based method for soils with different edaphic properties and microbial biomass

The suitability of a soil proteome analysis based on previous cell extraction by gradient centrifugation was tested in semiarid soils with distinct edaphic properties and microbial biomass after enrichment with carbon and nitrogen. A sandy loam soil with low organic carbon content reached higher microbial biomass (estimated by PLFAs) after stimulation with nutrient sources (glucose and proline) than a naturally rich soil. However, the extractability of soil microbial cells was higher in a poor soil with high electrical conductivity probably due to the high saline content. The number of identified proteins in the poor soils reached 71 with proteins related to energy processes, transport and nucleic acid metabolism representing the highest percentage. High organic carbon content negatively influences cell extraction and protein separation and analysis. Soil texture and/or salinity might be related to the expression of proteins involved in the removal of reactive oxygen species (ROS) such catalase and superoxide dismutase (SOD) under active metabolism and microbial biomass development     

Effect of liming on phosphate sorption by acid soils

The sorption of phosphate (P) by four strongly acid Fijian soils from 0.01 M CaCl₂ decreased with increasing pH up to pH 5.5–6.0 and then increased again. The initial decrease in P sorption with increasing pH appears to result from an interaction between added P, negative charge, and the electrostatic potential in the plane of sorption. The results of a sorption study, involving KCl or CaCl₂ of varying concentrations as the background electrolyte and using Nadroloulou soil incubated with KOH or Ca(OH)₂, suggested that the increase in P sorption at pH values > 6.0 was caused by the formation of insoluble Ca-P compounds. For some soils this is consistent with the results of an isotopic-exchange study in which incubation with lime caused marked reductions in the amounts of exchangeable P at high pH.     

Effect of reducing conditions on P sorption of soils

Abstract

The phosphate sorption (P _sor) capacity of soils increased when the soils were reduced (Willet and Higgins, Aust. J. Soil Res., 16, 319–326, 1978). The present study aimed at the elucidation of this mechanism using Na₂S₂O₄ and 5 different soils. The P _sor of the 5 soils increased with the addition of a small amount of Na₂S₂O₄. Fe(II) was released from the soils with the addition of the same small amount of Na₂S₂O₄. Furthermore, when the amount of FeCl₂ corresponding to the amount of Fe(II) released along with the small amount of Na₂S₂O₄ was added, the P _sor of the soil increased. However, the P _sor of the lowland soils, of which the hydrous Fe oxide content was lower than the others, decreased when the amount of Na₂S₂O₄ addition was increased up to 150–200 g kg^?1. Based on these results, the following process is inferred for the increase in the P _sor of the soils when they are reduced. Hydrous Fe oxide in soil takes the form of very fine, high-density particles and reacts with P mainly on their surface. When a small amount of Na₂S₂O₄ is added, the hydrous Fe oxide is partially reduced, dissolved and finally re-precipitates with P by oxidation with O₂ from the air during the experiment.     

Mineralization of carbon and nitrogen from cowpea leaves decomposing in soils with different levels of microbial biomass

Soils with greater levels of microbial biomass may be able to release nutrients more rapidly from applied plant material. We tested the hypothesis that the indigenous soil microbial biomass affects the rate of decomposition of added green manure. Cowpea (Vigna unguiculata L.) Walp.] leaves were added to four soils with widely differing microbial biomass C levels. C and N mineralization of the added plant material was followed during incubation at 30°C for 60 days. Low levels of soil microbial biomass resulted in an initially slower rate of decomposition of soil-incorporated green manure. The microbial biomass appeared to adjust rapidly to the new substrate, so that at 60 days of incubation the cumulative C loss and net N mineralization from decomposing cowpea leaves were not significantly affected by the level of the indigenous soil microbial biomass.     

Adenylates in the soil microbial biomass at different temperatures

Five soils from temperate sites (Germany; 2 arable and 3 grassland) were incubated aerobically at 5, 10, 15, 20, 25, 35, and 40 °C for 8 days. Soils were analysed for soil microbial biomass C, biomass N, AMP, ADP, and ATP to determine whether the increase in the ATP-to-microbial biomass C ratio with increasing temperature was either due to an increase in the adenylate energy charge (AEC) or de novo synthesis of ATP, or both. Around 80% of the variance in microbial biomass C and biomass N was explained by differences in soil properties, only 7% by the temperature treatments. Averaging the data of all 5 soils for each incubation temperature, the microbial biomass C content decreased with increasing temperature from 15 to 40 °C continuously by 2.5 μg g⁻¹ soil °C⁻¹ after 8-days' incubation. However, this decrease was not accompanied by a similar decrease in microbial biomass N. The average microbial biomass C/N ratio was 6.8. Between 54 and 76% of the variance in AMP, ADP, ATP and the sum of adenylates was explained by differences in soil properties and between 14 (ADP) and 27% (ATP) by the temperature treatments. However, temperature effects on AMP and ADP were variable and inconsistent. In contrast, ATP and consequently also the sum of adenylates increased continuously from 5 to 30 °C followed by a decline to 40 °C. The AEC showed similarly a small, but significant increase with increasing temperature from 0.73 to 0.85 at 30 °C. Consequently, the majority of the variance, i.e. roughly 60% in AEC values, but also in ATP-to-microbial biomass C ratios was explained by the incubation temperature. The mean ATP-to-microbial biomass C ratio increased from 4.7 μmol g⁻¹ at 5 °C to a 2.5 fold maximum of 12.0 μmol g⁻¹ at 35 °C. This increase was linear with a rate of 0.26 μmol ATP g⁻¹ microbial biomass C °C⁻¹. The energy for the extra ATP produced during temperature increase is probably derived from an accelerated turnover of endocellular C reserves in the microbial biomass.     

Activity and biomass of soil microorganisms at different depths

We measured microbial biomass C and soil organic C in soils from one grassland and two arable sites at depths of between 0 and 90 cm. The microbial biomass C content decreased from a maximum of 1147 (0–10 cm layer) to 24 g g^-1 soil (70–90 cm layer) at the grassland site, from 178 (acidic site) and 264 g g^-1 soil (neutral site) at 10–20 cm to values of between 13 and 12 g g^-1 soil (70–90 cm layer) at the two arable sites. No significant depth gradient was observed within the plough layer (0–30 cm depth) for biomass C and soil organic C contents. In general, the microbial biomass C to soil organic C ratio decreased with depth from a maximum of between 1.4 and 2.6% to a minimum of between 0.5 and 0.7% at 70–90 cm in the three soils. Over a 24-week incubation period at 25°C, we examined the survival of microbial biomass in our three soils at depths of between 0 and 90 cm without external substrate. At the end of the incubation experiment, the contents of microbial biomass C at 0–30 cm were significantly lower than the initial values. At depths of between 30 and 90 cm, the microbial biomass C content showed no significant decline in any of the four soils and remained constant up to the end of the experiment. On average, 5.8% of soil organic C was mineralized at 0–30 cm in the three soils and 4.8% at 30–90 cm. Generally, the metabolic quotient qCO₂ values increased with depth and were especially large at 70–90 cm in depth.     

Determination of carbon and nitrogen in microbial biomass of southern-Taiga soils by different methods

The results of methods for determining microbial biomass carbon vary in reproducibility among soils. The fumigation-extraction and substrate-induced respiration methods give similar results for Albic Luvisol and Gleyic Fluvisol, while the results of the rehydration method are reliably higher. In Histic Fluvisol, relatively similar results are obtained using the fumigation-extraction and rehydration methods, and the substrate-induced respiration method gives almost halved results. The seasonal dynamics of microbial biomass carbon also varies depending on the method used. The highest difference is typical for the warm period, when the concentrations found by the extraction and substrate-induced methods poorly agree between two out of three soils studied. The concentration of microbial biomass nitrogen is less sensitive to the analytical method: the differences between the results of the fumigation-extraction and rehydration methods are statistically insignificant in the all soils. To reveal stable relationships between the results of determining microbial carbon and the soil properties and analytical method, a large diversity of soils should be studied. This will allow for proposing of conversion factors for the recalculation of the obtained values to the concentrations of carbon and nitrogen in microbial biomass for different soils (or soil groups) and, hence, the more correct comparison of the results obtained by different methods.     

Plant litter decomposition and microbial characteristics in volcanic soils (Mt Etna,Sicily) at different stages of development

Soils at different developmental stages were sampled from eight sites on the slopes of Mt Etna, Sicily (Italy) and characterized for total C, microbial biomass and microbial respiration. The values of these parameters were greatest for the most developed soils, but differences in recent management and site characteristics limited analysis of trends with soil development across the eight sites. The decomposition kinetics of both intact leaf litter and the water-insoluble fraction of leaf litter from three common species on Etna [Etnean broom (Genista aetnensis), European chestnut (Castanea sativa), and Corsican pine (Pinus nigra)] were determined in four of the soils (the two with the smallest and the two with the largest organic C contents) in a laboratory experiment over 168 days to test two hypotheses. First, that the readily mineralized fraction of added plant C is greater when the plant material decomposes in well-developed soils compared to less developed soils, and second, that the microbial communities in less developed soils are less efficient at mineralizing C from low quality plant residues. The first hypothesis held for Genista and Pinus litter, but not Castanea litter. The second hypothesis was supported for the Castanea and Pinus litter, but not for the Genista litter. Thus, the general applicability of the hypotheses was dependent on the precise source and characteristics of the litter.     

Effects of equilibrating salt solutions on phosphate sorption by soils

Abstract

Several equilibrating salt solutions have been used in the studies of P sorption by soils and sediments. This study was conducted to evaluate the effects of 10 salt solutions on estimation of P sorption by soils. Results obtained showed that, when the equilibrating solution was made to contain 0.01M with respect to CaCl₂, Ca(NO₃)₂, CaSO₄, MgCl₂, KCl, LiCl, Nacl, or KHCO₃, the amount of P sorbed by soil always exceeded the amount sorbed from the soil‐water system. In comparison with the amount of P sorbed from water, 0.01M NaHCO₃ reduced P sorption by soils. Use of THAM buffer (0.05M pH 7.0) to control the pH increased P sorption by some soils and decreased P sorption by others, relative to that sorbed from the soil‐water system. The results indicated that inclusion of salts in the equilibrating solution for P‐sorption studies should be avoided, especially in studies related to water quality.     

Turnover of root-derived material and related microbial biomass formation in soils of different texture

Wheat plants were grown on two soils of different texture, a sandy soil and a silty clay loam, in an atmosphere containing ¹⁴CO₂. The ¹⁴C and total C content of the shoots, roots, soil rhizosphere CO₂ and soil microbial biomass were measured 21, 28, 35 and 42 days after germination. There was a pronounced effect of soil texture on the turnover of root-derived C through the microbial biomass. Turnover was relatively fast and at a constant rate in the sandy soil but slowed down in the clay soil, following an initial high assimilation of root products into the microbial biomass.Four percent of the total fixed ¹⁴C was retained in the clay loam after 6 weeks compared with a corresponding value of 1.2% for the sandy soil. The proportion of fixed ¹⁴C recovered as rhizosphere CO₂ at each of the sampling times was relatively constant for the sandy soil (ca 19%) but decreased from 17% at day 28 to 11% at day 42 in the clay soil. The proportion of total fixed ¹⁴C in the soil biomass as measured by a fumigation technique increased to a maximum value of 20% after 6 weeks in the sandy soil but decreased in the clay soil from 86% at day 21 to 26% after 42 days plant growth.