Sorption and desorption of glyphosate,MCPA and tetracycline and their mixtures in soil as influenced by phosphate

Phosphate fertilizers and herbicides such as glyphosate and MCPA are commonly applied to agricultural land, and antibiotics such as tetracycline have been detected in soils following the application of livestock manures and biosolids to agricultural land. Utilizing a range of batch equilibrium experiments, this research examined the competitive sorption interactions of these chemicals in soil. Soil samples (0-15 cm) collected from long-term experimental plots contained Olsen P concentrations in the typical (13 to 20 mg kg^?1) and elevated (81 to 99 mg kg^?1) range of build-up phosphate in agricultural soils. The elevated Olsen P concentrations in field soils significantly reduced glyphosate sorption up to 50%, but had no significant impact on MCPA and tetracycline sorption. Fresh phosphate additions in the laboratory, introduced to soil prior to, or at the same time with the other chemical applications, had a greater impact on reducing glyphosate sorption (up to 45%) than on reducing tetracycline (up to 13%) and MCPA (up to 8%) sorption. The impact of fresh phosphate additions on the desorption of these three chemicals was also statistically significant, but numerically very small namely < 1% for glyphosate and tetracycline and 3% for MCPA. The presence of MCPA significantly reduced sorption and increased desorption of glyphosate, but only when MCPA was present at concentrations much greater than environmentally relevant and there was no phosphate added to the MCPA solution. Tetracycline addition had no significant effect on glyphosate sorption and desorption in soil. For the four chemicals studied, we conclude that when mixtures of phosphate, herbicides and antibiotics are present in soil, the greatest influence of their competitive interactions is phosphate decreasing glyphosate sorption and the presence of phosphate in solution lessens the potential impact of MCPA on glyphosate sorption. The presence of chemical mixtures in soil solution has an overall greater impact on the sorption than desorption of individual organic chemicals in soil.     

Degradation of herbicides in two sandy aquifers under different redox conditions.

We examined the potential for complete degradation (mineralisation) of the four [ring-U-14C]herbicides mecoprop, isoproturon, atrazine, and metsulphuron-methyl in two sandy aquifers representing aerobic, denitrifying, sulphate-reducing, and methanogenic conditions. Slurries with sediment and groundwater were set-up aerobically or anaerobically in the presence of the electron-acceptor prevailing at the sampling site, amended with 25 microg l(-1) herbicide, and incubated at 10 degrees C. Considerable mineralisation was only observed in sediment from the plough layer incubated aerobically. Here, 30% of 14C-mecoprop was recovered as 14CO2 after 15 days and 15% of isoproturon was recovered as 14CO2 after 267 days. Only 7% of mecoprop was recovered as 14CO2 after 313 days in sediment from the aquifer below sampled at 1.95-3.00 mbs (m below the surface). In denitrifying and methanogenic slurries, 3% of 14C added as mecoprop was recovered as 14CO2. Isoproturon was not mineralised except in the aerobic plough layer, and atrazine and metsulphuron-methyl were not mineralised under any of the conditions applied.     

Variation of MCPA, metribuzine, methyltriazine-amine and glyphosate degradation, sorption, mineralization and leaching in different soil horizons

Pesticide mineralization and sorption were determined in 75 soil samples from 15 individually drilled holes through the vadose zone along a 28 km long transect of the Danish outwash plain. Mineralization of the phenoxyacetic acid herbicide MCPA was high both in topsoils and in most subsoils, while metribuzine and methyltriazine-amine was always low. Organic matter and soil pH was shown to be responsible for sorption of MCPA and metribuzine in the topsoils. The sorption of methyltriazine-amine in topsoil was positively correlated with clay and negatively correlated with the pH of the soil. Sorption of glyphosate was tested also high in the subsoils. One-dimensional MACRO modeling of the concentration of MCPA, metribuzine and methyltriazine-amine at 2 m depth calculated that the average concentration of MCPA and methyltriazine-amine in the groundwater was below the administrative limit of 0.1 μg/l in all tested profiles while metribuzine always exceeded the 0.1 μg/l threshold value.     

Mineralisation studies of 14C-labelled metsulfuron-methyl, tribenuron-methyl, chlorsulfuron and thifensulfuron-methyl in one Danish soil and groundwater sediment profile.

Bacterial mineralisation of four sulfonylurea herbicides at 20 microg kg(-1) in a sandy soil from nine different depths in a sandy soil horizon (5-780 cm) was investigated in laboratory studies. Metsulfuron-methyl, chlorsulfuron, and tribenuron-methyl were 14C-labelled in the sulfonamide ring, while thifensulfuron-methyl was labelled in the thiophene ring. The highest mineralised amount in 126 days was observed for metsulfuron-methyl (40%) followed by tribenuron-methyl (25%), and thifensulfuron-methyl (11%). Chlorsulfuron showed low mineralisation in all the soils tested (<4%). Mineralisation of the herbicides metsulfuron-methyl and tribenuron-methyl varied according to soil depth (upper profile: 5-70 cm, and lower profile: 165-780 cm) and were proven faster in soil taken from depths 5-7 and 30-35 cm, and slower in depths 45-50 and 70-75 cm. Mineralisation was absent in the lower profile (165-780 cm). As an indicator of microbial activity bacterial counts were taken at the experimental start; these counts grouped in three levels: highest in the surface layer (5-7 cm), slightly lower in the depths 30-75 cm, and lowest in the lower profile (165-780 cm). Residual concentrations of metsulfuron-methyl correlated to the accumulated amount mineralised, with high residual concentrations in soil showing low mineralisation. Also chlorsulfuron showed high residual concentrations with increasing depth in the upper profile, but the relatively high dissipation at 30-35 cm and lower one at 45-50 cm could not be related with the lack of mineralisation. This shows that hydrolysis occurs, but mineralisation of the chloro-substituted sulfonamide is restricted. Tribenuron-methyl and thifensulfuron-methyl could not be detected due to interference with other compounds.     

Biomineralisation of 1,2,4-trichlorobenzene in soils by an adapted microbial population

In laboratory experiments the mineralisation of 14C-labelled 1,2,4-trichlorobenzene (1,2,4-TCB) in soils was studied by direct measurement of the evolved 14CO2. The degradation capacity of the indigenous microbial population was investigated in an agricultural soil and in a soil from a contaminated site. Very low mineralisation of 1% within 23 days was measured in the agricultural soil. Whereas in the soil from the contaminated site the mineralisation occurred very fast and in high rates; up to 62% of the initially applied amount of 1,2,4-TCB were mineralised within 23 days. The transfer of the adapted microbial population into the agricultural soil significantly enhanced the mineralisation of 1,2,4-TCB in this soil, reflecting, that the transferred microbial population survived and maintained its degradation ability in the new microbial ecosystem. Additional nutrition sources ((NH4)2HPO4) increased the mineralisation rates in the first days significantly in the contaminated soil. In the soil from the contaminated site high amounts of non extractable 14C-residues were formed.     

Degradation and sorption of imidacloprid in dissimilar surface and subsurface soils

Degradation and sorption/desorption are important processes affecting the leaching of pesticides through soil. This research characterized the degradation and sorption of imidacloprid (1-[(6-chloro-3-pyridinyl)-methyl]-N-nitro-2-imidazolidinimine) in Drummer (silty clay loam) and Exeter (sandy loam) surface soils and their corresponding subsurface soils using sequential extraction methods over 400 days. By the end of the incubation, approximately 55% of imidacloprid applied at a rate of 1.0 mg kg(-1) degraded in the Exeter sandy loam surface and subsurface soils, compared to 40% of applied imidacloprid within 300 days in Drummer surface and subsurface soils. At the 0.1 mg kg(-1) application rate, dissipation was slower for all four soils. Water-extractable imidacloprid in Exeter surface soil decreased from 98% of applied at day 1 to >70% of the imidacloprid remaining after 400 d, as compared to 55% in the Drummer surface soil at day 1 and 12% at day 400. These data suggest that imidacloprid was bioavailable to degrading soil microorganisms and sorption/desorption was not the limiting factor for biodegradation. In subsurface soils > 40% of (14)C-benzoic acid was mineralized over 21 days, demonstrating an active microbial community. In contrast, cumulative (14)CO(2) was less than 1.5% of applied (14)C-imidacloprid in all soils over 400 d. Qualitative differences in the microbial communities appear to limit the degradation of imidacloprid in the subsurface soils.     

Sorption, desorption, and degradation of (4-chloro-2-methylphenoxy)acetic acid in representative soils of the Danubian Lowland, Slovakia

Herbicide leaching through soil into groundwater greatly depends upon sorption-desorption and degradation phenomena. Batch adsorption, desorption and degradation experiments were performed with acidic herbicide MCPA and three soil types collected from their respective soil horizons. MCPA was found to be weakly sorbed by the soils with Freundlich coefficient values ranging from 0.37 to 1.03 mg^1−1/n kg⁻¹ L^1/n. It was shown that MCPA sorption positively correlated with soil organic carbon content, humic and fulvic acid carbon contents, and negatively with soil pH. The importance of soil organic matter in MCPA sorption by soils was also confirmed by performing sorption experiments after soil organic matter removal. MCPA sorption in these treated soils decreased by 37-100% compared to the original soils. A relatively large part of the sorbed MCPA was released from soils into aqueous solution after four successive desorption steps, although some hysteresis occurred during desorption of MCPA from all soils. Both sorption and desorption were depth-dependent, the A soil horizons exhibited higher retention capacity of the herbicide than B or C soil horizons. Generally, MCPA sorption decreased in the presence of phosphate and low molecular weight organic acids. Degradation of MCPA was faster in the A soil horizons than the corresponding B or C soil horizons with half-life values ranging from 4.9 to 9.6 d in topsoils and from 11.6 to 23.4 d in subsoils.     

Degradation and sorption of imidacloprid in dissimilar surface and subsurface soils

Degradation and sorption/desorption are important processes affecting the leaching of pesticides through soil. This research characterized the degradation and sorption of imidacloprid (1-[(6-chloro-3-pyridinyl)-methyl]-N-nitro-2-imidazolidinimine) in Drummer (silty clay loam) and Exeter (sandy loam) surface soils and their corresponding subsurface soils using sequential extraction methods over 400 days. By the end of the incubation, approximately 55% of imidacloprid applied at a rate of 1.0 mg kg^?1 degraded in the Exeter sandy loam surface and subsurface soils, compared to 40% of applied imidacloprid within 300 days in Drummer surface and subsurface soils. At the 0.1 mg kg^?1 application rate, dissipation was slower for all four soils. Water-extractable imidacloprid in Exeter surface soil decreased from 98% of applied at day 1 to > 70% of the imidacloprid remaining after 400 d, as compared to 55% in the Drummer surface soil at day 1 and 12% at day 400. These data suggest that imidacloprid was bioavailable to degrading soil microorganisms and sorption/desorption was not the limiting factor for biodegradation. In subsurface soils > 40% of ¹⁴C-benzoic acid was mineralized over 21 days, demonstrating an active microbial community. In contrast, cumulative ¹⁴CO₂ was less than 1.5% of applied ¹⁴C-imidacloprid in all soils over 400 d. Qualitative differences in the microbial communities appear to limit the degradation of imidacloprid in the subsurface soils.     

Adsorption and desorption processes of MCPA in Polish mineral soils

Studies on the adsorption and desorption of MCPA (4-chloro-2-methylophenoxyacetic acid) were performed in soil horizons of three representative Polish agricultural soils. The Hyperdystric Arenosol, the Haplic Luvisol and the Hypereutric Cambisol were investigated in laboratory batch experiments. Initially, both the adsorption and desorption proceeded rapidly, and either the equilibrium was reached after approximately 30 min or the process slowed down and continued at a slow rate. In the latter case, the equilibrium was reached after 8 hours. Data on the adsorption/desorption kinetics fitted well to the two-site kinetic model. The measured sorption and desorption isotherms were of L-type. The sorption distribution coefficients (K(ads) (d)) were in the range of 0.75--0.97 for Ap soil horizons and significantly lower in deeper soil layers. The corresponding desorption coefficients (K(des) (d)) were higher and ranged from 1.02 to 2.01. Both the adsorption and desorption of MCPA in all soil horizons was strongly and negatively related to soil pH. It appears that hydrophobic sorption plays a dominant role in the MCPA retention in topsoils whereas hydrophilic sorption of MCPA anions is the dominant adsorption mechanism in subsoils.     

Sorption of 3,4-dichloroaniline on four contrasting Greek agricultural soils and the effect of liming

Sorption of 3,4-dichloroaniline (3,4-DCA) on four typical Greek agricultural soils, with distinct texture, organic matter content and cation exchange capacities, was compared by using sorption isotherms and the parameters calculated from the fitted Freundlich equations. The sorption process of 3,4-DCA to the soil was completed within 48-72 h. The 3,4-DCA sorption on all soils was well described by the Freundlich equation and all sorption isotherms were of the L-type. The sandy clay loam soil with the highest organic matter content and a slightly acidic pH was the most sorptive, whereas the two other soil types, a high organic matter and neutral pH clay and a low organic matter and acidic loam, had an intermediate sorption capacity. A typical calcareous soil with low organic matter had the lowest sorption capacity which was only slightly higher than that of river sand. The 3,4-DCA sorption correlated best to soil organic matter content and not to clay content or cation exchange capacity, indicating the primary role of organic matter. The distribution coefficient (K(d)) decreased with increasing initial 3,4-DCA concentration and the reduction was most pronounced with the highly sorptive sandy clay loam soil, suggesting that the available sorption sites of the soils are not unlimited. Liming of the two acidic soils (the sandy clay loam and the loam) raised their pH (from 6.2 and 5.3, respectively) to 7.8 and reduced their sorption capacity by about 50 %, indicating that soil pH may be the second in importance factor (after organic matter) determining 3,4-DCA sorption.     

Aminocyclopyrachlor sorption-desorption and leaching from three Brazilian soils

This study aimed to evaluate the sorption-desorption and leaching of aminocyclopyrachlor from three Brazilian soils. The sorption-desorption of ¹⁴C-aminocyclopyrachlor was evaluated using the batch method and leaching was assessed in glass columns. The Freundlich model showed an adequate fit for the sorption-desorption of aminocyclopyrachlor. The Freundlich sorption coefficient [K_{f (sorption)}] ranged from 0.37 to 1.34 µmol ^(1–1/n) L^1/n kg^?1 and showed a significant positive correlation with the clay content of the soil, while the K_{f (desorption)} ranged from 3.62 to 5.36 µmol ^(1–1/n) L^1/n kg^?1. The K_{f (desorption)} values were higher than their respective K_{f (sorption)}, indicating that aminocyclopyrachlor sorption is reversible, and the fate of this herbicide in the environment can be affected by leaching. Aminocyclopyrachlor was detected at all depths (0?30 cm) in all the studied soils, where leaching was influenced by soil texture. The total herbicide leaching from the sandy clay and clay soils was <0.06%, whereas, ～3% leached from the loamy sand soil. The results suggest that aminocyclopyrachlor has a high potential of leaching, based on its low sorption and high desorption capacities. Therefore, this herbicide can easily contaminate underground water resources.     

Carbendazim sorption-desorption in Vietnamese soils

Four Vietnamese soils (denoted AG, CT, ST and TG) which differed with respect to pH (pH 2.9-5.4), clay (17-50%) and organic matter (0.3-9.8%) content, were selected for sorption and desorption studies of carbendazim using the batch equilibration technique. Sorption increased with increasing organic carbon (OC) and clay content. Kd values for carbendazim sorption on AG, CT, ST, TG soils at initial concentration of 20 microg/g were 12.5, 127, 8.1 and 9.6 ml/g, respectively. The OC partition coefficients (Koc) for AG, CT, ST and TG were 1140, 1300, 2700 and 960 ml/g, respectively. Carbendazim was strongly sorbed and the binding was less reversible in the acid sulfate soil (CT), than in the other soils. The CT soil had both the highest OC content (9.8%) and the highest clay content (49.8%). The influence of pH on carbendazim sorption was studied in the ST and CT soils. Sorption of carbendazim by the sandy ST soil (OC 0.3%; clay content 26.3%) increased as the pH decreased, while sorption of carbendazim by the CT soil decreased as pH decreased.     

Sorption of 3,4-dichloroaniline on four contrasting Greek agricultural soils and the effect of liming

Sorption of 3,4-dichloroaniline (3,4-DCA) on four typical Greek agricultural soils, with distinct texture, organic matter content and cation exchange capacities, was compared by using sorption isotherms and the parameters calculated from the fitted Freundlich equations. The sorption process of 3,4-DCA to the soil was completed within 48–72 h. The 3,4-DCA sorption on all soils was well described by the Freundlich equation and all sorption isotherms were of the L-type. The sandy clay loam soil with the highest organic matter content and a slightly acidic pH was the most sorptive, whereas the two other soil types, a high organic matter and neutral pH clay and a low organic matter and acidic loam, had an intermediate sorption capacity. A typical calcareous soil with low organic matter had the lowest sorption capacity which was only slightly higher than that of river sand. The 3,4-DCA sorption correlated best to soil organic matter content and not to clay content or cation exchange capacity, indicating the primary role of organic matter. The distribution coefficient (K _d) decreased with increasing initial 3,4-DCA concentration and the reduction was most pronounced with the highly sorptive sandy clay loam soil, suggesting that the available sorption sites of the soils are not unlimited. Liming of the two acidic soils (the sandy clay loam and the loam) raised their pH (from 6.2 and 5.3, respectively) to 7.8 and reduced their sorption capacity by about 50 %, indicating that soil pH may be the second in importance factor (after organic matter) determining 3,4-DCA sorption.     

Studies on the sorption and desorption characteristics of Zn(II) on the surface soils of nuclear power plant sites in India using a radiotracer technique

Zinc adsorption was studied in the soils of three nuclear power plant sites of India. 65Zn was used as a radiotracer to study the sorption characteristics of Zn(II). The sorption of zinc was determined at 25 and 45 degrees C at pH 7.8+/-0.2 in the solution of 0.01 M Ca(NO3)2 as supporting electrolyte. The sorption data was tested both in Freundlich and Langmuir isotherms and could be described satisfactorily. The effect of organic matter and other physico-chemical properties on the uptake of zinc was also studied in all the soil samples. The results showed that the cation exchange capacity, organic matter, pH and clay content were the main contributors to zinc sorption in these soils. The adsorption maximum was found to be higher in the soil on Kakarpara Atomic Power Plant sites soils having high organic matter and clay content. The zinc supply parameters of the soils are also discussed. In the desorption studies, the sequential extraction of the adsorbed zinc from soils showed that the diethylene triamine penta acetic acid extracted maximum amount of adsorbed zinc than CaCl2 and Mg(NO3)2. The zinc sorption on the soil and amount of zinc retention after extractants desorption shows a positively correlation with vermiculite and smectite mineral content present in the clay fraction of the soil. The amount desorbed by strong base (NaOH) and demineralised water was almost negligible from soils of all the sites, whereas the desorption by strong acid (HNO3) was 75-96% of the adsorbed zinc.     

Sorption–desorption of imidacloprid onto a lacustrine Egyptian soil and its clay and humic acid fractions

Sorption–desorption of the insecticide imidacloprid 1-[(6-chloro-3-pyridinyl)-methyl]-N-nitro-2-imidazolidinimine onto a lacustrine sandy clay loam Egyptian soil and its clay and humic acid (HA) fractions was investigated in 24-h batch equilibrium experiments. Imidacloprid (IMDA) sorption–desorption isotherms onto the three sorbents were found to belong to a non-linear L-type and were best described by the Freundlich model. The value of the IMDA adsorption distribution coefficient, Kd_ads, varied according to its initial concentration and was ranged 40–84 for HA, 14–58 for clay and 1.85–4.15 for bulk soil. Freundlich sorption coefficient, Kf_ads, values were 63.0, 39.7 and 4.0 for HA, clay and bulk soil, respectively. The normalized soil Koc value for imidacloprid sorption was ～800 indicating its slight mobility in soils. Nonlinear sorption isotherms were indicated by 1/n_ads values <1 for all sorbents. Values of the hysteresis index (H) were <1, indicating the irreversibility of imidacloprid sorption process with all tested sorbents. Gibbs free energy (ΔG) values indicated a spontaneous and physicosorption process for IMDA and a more favorable sorption to HA than clay and soil. In conclusion, although the humic acid fraction showed the highest capacity and affinity for imidacloprid sorption, the clay fraction contributed to approximately 95% of soil-sorbed insecticide. Clay and humic acid fractions were found to be the major two factors controlling IMDA sorption in soils. The slight mobility of IMDA in soils and the hysteresis phenomenon associated with the irreversibility of its sorption onto, mainly, clay and organic matter of soils make its leachability unlikely to occur.     

Mineralisation of target hydrocarbons in three contaminated soils from former refinery facilities

This study investigated the microbial degradation of ¹⁴C-labelled hexadecane, octacosane, phenanthrene and pyrene and considered how degradation might be optimised in three genuinely hydrocarbon-contaminated soils from former petroleum refinery sites. Hydrocarbon mineralisation by the indigenous microbial community was monitored over 23 d. Hydrocarbon mineralisation enhancement by nutrient amendment (biostimulation), hydrocarbon degrader addition (bioaugmentation) and combined nutrient and degrader amendment, was also explored. The ability of indigenous soil microflora to mineralise ¹⁴C-target hydrocarbons was appreciable; ≥16% mineralised in all soils. Generally, addition of nutrients or degraders increased the rates and extents of mineralisation of ¹⁴C-hydrocarbons. However, the addition of nutrients and degraders in combination had a negative effect upon ¹⁴C-octacosane mineralisation and resulted in lower extents of mineralisation in the three soils. In general, the rates and extents of mineralisation will be dependent upon treatment type, nature of the contamination and adaptation of the ingenious microbial community.     

Mineralisation of 2,6-dichlorobenzamide (BAM) in dichlobenil-exposed soils and isolation of a BAM-mineralising Aminobacter sp

Mineralisation of the groundwater contaminant 2,6-dichlorobenzamide (BAM), a metabolite from the herbicide 2,6-dichlorobenzonitrile (dichlobenil), was studied in soil samples obtained from 39 locations previously exposed to dichlobenil. Rapid BAM mineralisation was detected in samples from six locations with 5.2-64.6% of the added BAM mineralised within 48-50 days. From one location rapid BAM mineralisation was observed in soil samples down to a depth of 2 m below the surface. One location with fast BAM mineralisation showed significant dichlobenil degradation activity with 25.5% of the added dichlobenil being mineralised within 50 days. By inoculating soil showing the fastest mineralisation of BAM into a mineral medium with BAM as the only carbon and nitrogen source an enrichment culture was established. Community analysis based on extracted DNA revealed a change of the bacterial community but without any clear indication of key members within the BAM-mineralising culture. Parallel cultivation resulted for the first time in the isolation of a BAM-mineralising bacterium, identified as an Aminobacter sp.     

The influence of organic matter on sorption and fate of glyphosate in soil - Comparing different soils and humic substances

Soil organic matter (SOM) is generally believed not to influence the sorption of glyphosate in soil. To get a closer look on the dynamics between glyphosate and SOM, we used three approaches: I. Sorption studies with seven purified soil humic fractions showed that these could sorb glyphosate and that the aromatic content, possibly phenolic groups, seems to aid the sorption. II. Sorption studies with six whole soils and with SOM removed showed that several soil parameters including SOM are responsible for the strong sorption of glyphosate in soils. III. After an 80 day fate experiment, ∼40% of the added glyphosate was associated with the humic and fulvic acid fractions in the sandy soils, while this was the case for only ∼10% of the added glyphosate in the clayey soils. Glyphosate sorbed to humic substances in the natural soils seemed to be easier desorbed than glyphosate sorbed to amorphous Fe/Al-oxides.     

Appendix

Abstract

The interaction of glyphosate [N‐(phosphonomethyl)‐glycine] with four typical European soils is reported. Results of adsorption and desorption isotherms show that the interaction of glyphosate with these soils was mainly related to content of iron and aluminium amorphus hydroxides. Moreover, it was found that the presence of divalent cations in 2: 1 clay minerals also contribute to glyphosate adsorption. The S‐type form of the adsorption isotherms revealed the existence of two different binding sites. These were exchangeable cations at low herbicide concentration and Fe and Al at higher glyphosate concentrations. The K maximum values of adsorption provided by the linear form of the Langmuir equation were found to be more consistent with soil parameters than those calculated by the Freundlich equation. The order of desorption from the soils was the reverse of that found for adsorption. Moreover, desorption varied from around 15 to 80% of the adsorbed herbicide according to the soil characteristics. This indicated that glyphosate adsorption on soils is far from being permanent and leaching to lower soil horizons with limited biological activity may occur.     

Sorption and mobility of metronidazole, olaquindox, oxytetracycline and tylosin in soil

Laboratory studies were conducted to characterise four different antibiotic compounds with regard to sorption and mobility in various soil types. Distribution coefficients (Kd values) determined by a batch equilibrium method varied between 0.5 and 0.7 for metronidazole, 0.7 and 1.7 for olaquindox and 8 and 128 for tylosin. Tylosin sorption seems to correlate positively with the soil clay content. No other significant interactions between soil characteristics and sorption were observed. Oxytetracycline was particularly strongly sorbed in all soils investigated, with Kd values between 417 in sand soil and 1026 in sandy loam, and no significant desorption was observed. Soil column leaching experiments indicated large differences in the mobility of the four antibiotic substances, corresponding to their respective sorption capabilities. For the weakly adsorbed substances metronidazole and olaquindox the total amounts added were recovered in the leachate of both sandy loam and sand soils. For the strongly adsorbed oxytetracyline and tylosin nothing was detected in the leachate of any of the soil types, indicating a much lower mobility. Results from defractionation and extraction of the columns (30 cm length) showed that 60-80% of the tylosin added had been leached to a depth of 5 cm in the sandy loam soil and 25 cm in the sand soil.