Protection of Nitrosomonas europaea colonizing clay minerals from inhibition by nitrapyrin

Transcription

1 Journal of' General Microbiology (199 l), 17, Printed in Great Britain 192 Protection of Nitrosomonas europaea colonizing clay minerals from inhibition by nitrapyrin S. J. POWLL? and J. I. PROSSR" Department of Molecular and Cell Biology, University of Aberdeen, Marischal College, Aberdeen AB9 IAS, UK (Received 29 April 1991 ; accepted 2 May 1991) Nitrite production by Nifrosomonas europaea in inorganic liquid medium containing ammonium was limited by reduction in ph. In the presence of montmorillonite and vermiculite, expanding clays with high cation-exchangecapacity (CC), nitrite yield was increased, ammonia oxidation continued at ph values below those which inhibited growth in the absence of clays and growth was biphasic. The first phase was similar to that in the absence of clays, while the second was characterized by a lower rate of nitrite production. Illite, a non-expanding clay with low CC, had no significant effect on ammonia oxidation, while oxidation of ammonia-treated vermiculite (ATV) occurred with no significant change in the ph of the medium. ATV, montmorillonite and vermiculite, but not illite, protected cells from inhibition by nitrapyrin at concentrations inhibitory to cells growing in suspended culture. This protection was maintained in ATV homo-ionic to A1+, but montmorillonite made homo-ionic to A1+ did not provide protection from inhibition. Attachment of cells to clays with high CC is therefore advantageous in providing exchange at the clay surface of NH,+ and H+ produced by ammonia oxidation, in reducing ph toxicity, and in protecting cells from inhibition. Introduction Autotrophic nitrification is the oxidation of ammonia, via nitrite, to nitrate, and in agricultural soils can lead to significant losses of ammonia-based nitrogen fertilizers through leaching and denitrification of nitrate. In addition, there is increasing concern regarding high levels of nitrate in run-off and groundwaters and the consequent potential risks of cancer through formation of nitrosamines and methaemoglobinaemia following consumption of nitrate in drinking water. Nitrification is carried out in two stages, by autotrophic ammonia- and nitrite-oxidizing bacteria. Oxidation of ammonia is usually slower than that of nitrite, which rarely accumulates in soil. Because of this, and the phytotoxicity and mobility of nitrite, the development of commercial inhibitors of nitrification has concentrated on specific inhibitors of ammonia oxidation, the most extensively studied and widely used being nitrapyrin (2-chloro-6- trichloromethyl pyridine), first reported by Goring (1962). The efficacy of nitrapyrin and other nitrification inhibitors depends on both physicochemical and biolo- t Present address: Malthus Instruments Ltd, The Manor, Manor Royal, Crawley, West Sussex RHlO 2PY, UK. Abbreviations: ATV, ammonia-treated vermiculite ; CC, cation exchange capacity; NP, nitrapyrin. gical factors (Keeney, 1986) and the sensitivity of nitrifiers to nitrapyrin (Rodgers & Ashworth, 1982), potassium ethyl xanthate (Underhill & Prosser, 1985) and herbicides (Ratnayake & Audus, 1978) has been shown to be greater in liquid culture than in soil. Protection of nitrification from inhibition by nitrapyrin in soil has been attributed to adsorption and inactivation of the inhibitor by organic matter, chemical hydrolysis, ph and temperature effects, and selection of resistant strains (see Powell, 1986). The importance of the last of these explanations has been questioned by Powell & Prosser (1986), who demonstrated that concentrations of nitrapyrin giving inhibition in sterilized soil inoculated with a pure culture of Nitrosornonas europaea were an order of magnitude greater than those giving equivalent inhibition of the same organism in liquid culture. Although the effects of soil organic matter content on inhibition have been investigated (see Keeney, 1986), little attention has been paid to the role of clay minerals, the other major component of soil. Their effects on microbial processes have been reviewed by Burns (198) and include the stimulation of activity, protection from inhibition and phage attachment, and increased microbial survival. ffects of clays on nitrification have rarely been investigated but Sims & Little (197) found stimulation of nitrification by clinoptilite in an activated O 1991 SGM

2 1924 S. J. Powell and J. I. Prosser sludge process. Kunc & Stotzky (198) and Macura & Stotzky (1 98) found increased nitrification in soils containing higher levels of montmorillonite, while kaolinite had no detectable effect. Bozian & Stotzky (1976) also reported protection by montmorillonite from inhibition of nitrification by sulphur dioxide. In these, and most other studies, the effects of clay minerals cannot be isolated from those of other soil components and microbial populations. One exception is the study of the oxidation by N. europaea of ammonia adsorbed onto vermiculite (Armstrong & Prosser, 1988), which provided evidence of a mechanism for nitrification in acidic environments through localized exchange of H+ and NH,+ ions at the clay surface. Here we describe ammonia oxidation by pure cultures of N. europaea in the presence of this and other clay minerals and their role in the protection of nitrifiers from inhibition by nitrapyrin. Methods Organism and growth medium. A strain of Nitrosomonas europaea, derived from a soil isolate kindly supplied by Dr R. M. Macdonald (Rothamsted xperimental Station, Harpenden, UK), was used in all experiments. The strain, designated N. europaea sp.1 in a previous publication (Powell & Prosser, 1986), arose during continued subculturing and differed from the parent strain with regard to maximum specific growth rate and sensitivity to specific inhibitors of ammonia oxidation. Properties of both strains are described elsewhere (Powell & Prosser, 1986) and the derived strain is referred to here as N. europaea. Routine growth and maintenance were carried out in inorganic liquid medium (SW) (Skinner & Walker, 1961) as described by Keen & Prosser (1987). Clay minerals. All clay minerals were kindly provided by Dr J. Russell (Macaulay Land Use Research Institute, Aberdeen). Montmorillonite no. 2 (original source Chambers, Arizona, USA) and illite (original source Fifthian) had a particle size range of mm. Vermiculite and ammonia-treated vermiculite (ATV) were hydrobiotites (1 : 1, mica :vermiculite) supplied in granular form by Mandoval but are referred to as vermiculite because of the interaction between the vermiculite layers and ammonia (Ahlrichs et al., 1972). ATV was prepared by passing anhydrous ammonia through vermiculite (Russell & Fraser, 1977) and contained 12.6 mg N (g clay)-'. Both vermiculite and ATV had particle sizes in the range mm. When required, montmorillonite and ATV were made homo-ionic to aluminium by the method of Santoro & Stotzky (1967). Clay particles (5 g) were shaken in 5 ml 1.5 M-AICI,.6H2 in 1 ml rlenmeyer flasks for 1 h. Clays were separated by centrifugation and the procedure was repeated three times before four 1 h washings in deionized water and final air drying. Preparation and hydrolysis of nitrapyrin. Nitrapyrin was supplied by Dow Chemical Co. in the form of N-Serve 24 emulsion. Inhibitor solutions were prepared by emulsification in.5% ethanol in water followed by dilution in distilled water and filter-sterilization prior to addition to the culture media. The effect of ATV on the rate of hydrolysis was determined in triplicate in 25 ml rlenmeyer flasks containing 15 ml sterile SW plus nitrapyrin (final concentration.5 pg ml-l), with and without addition of.75g ATV, incubated with shaking at C. Nitrapyrin was extracted from 1ml samples by shaking in 5 ml.5 M-sodium bicarbonate and 1 ml hexane (Analar grade, redistilled). xtracts were analysed using a Packard 49 gasliquid chromatogram with a 6Ni electron capture detector (Briggs, 1975). The column (length.5 m; internal diameter,.46 cm) was packed with 5% QFI on Chromosorb W, AW-DCMS, 8-1 mesh (Phase Separations). The carrier gas was argon/methane at 4 kpa with oven, detector and injector at 128, 25 and 15 "C, respectively. The retention time for nitrapyrin was 1 min and the lower limit of detection for a 1 pl sample injection was 2 pg ml-'. Linear regression was used to calculate the first-order rate constant for hydrolysis of nitrapyrin. Growth studies. rlenmeyer flasks (25 ml) containing 15 ml SW, with an initial ammonium concentration of 5 pg NHZ-N ml-l, were inoculated with 2 ml of a stationary phase culture of N. europaea. All clay minerals, except ATV, were added at a concentration of 5 mg ml-i to the complete medium. ATV was added at the same concentration, but to ammonia-free SW medium, giving a final ammonium concentration equivalent to 55 pg NH,+-N m1-i. All treatments, and uninoculated controls, were prepared in triplicate. Flasks were autoclaved at 121 "C for 15 min and the ph was adjusted to 8 by aseptic addition of sterile 5% (w/v) sodium carbonate. After inoculation, flasks were incubated at "C, shaken at 15 r.p.m., and samples were removed at regular intervals for determination of ph and for analysis of soluble ammonium and nitrite using an Autoanalyser I1 System (Technicon Instruments). Nitrapyrin was added at a final concentration of.5 or 1. pg ml-i to mid-exponential phase cultures, assessed as those cultures in which approximately NOT-N ml-i had been produced. Growth was also studied in the presence of.75g ATV contained within 4 cm lengths of visking tubing, previously boiled for 1 h in distilled water. Tubing containing ATV was closed by nylon thread, autoclaved in 25 ml rlenmeyer flasks containing 15 ml ammonia-free SW medium, inoculated and incubated as described above. ffects of different treatments were assessed statistically by comparing means of specific growth rates from triplicate flasks using the Student's t-test. Results Influence of montmorillonite, vermiculite and illite on ammonia oxidation In the absence of clay minerals, ammonia oxidation continued until the ph of the medium fell to 6.8 (Fig. 1 a). Nitrite concentration increased exponentially at a constant specific rate until the ph decreased below 7-5, allowing calculation of a specific growth rate value of -45 h-i. As the ph decreased below 7.5, the specific growth rate decreased and the final nitrite concentration was 4.5 pg NO,-N ml-l. Cessation of growth was not due to inhibition by nitrite as complete conversion of 5 pg NH,+-N ml-1 occurred in ph-controlled cultures with no reduction in the specific growth rate (Powell, 1985). Biphasic growth occurred in the presence of montmorillonite (Fig. 1 b). During the first phase, which ended at 8 h, the specific growth rate was.8 h-l, and was significantly less than that of control cultures (P =.). The specific growth rate then fell to.26 h-l, significantly less than that in the first phase (P =.1). The change in the specific growth rate occurred at a nitrite concentration of approximately 2 pg ml-l but montmorillonite buffered the medium and exponential

3 Inhibition of clay-colonizing nitr$ers I I I I I I I I I I Fig. 1. Nitrite concentration () and ph () during growth of N. europuea in 15 ml inorganic medium containing (a) no clays (6) montmorillonite (.5 g l-'), (c) vermiculite (.5 g 1-l) or (d) illite (.5 g 1-l). For nitrite data, straight lines indicate regression lines of semi-logarithmic plots of nitrite concentration used to calculate the specific growth rates during the first and, when present, the second phases of growth. growth continued for 125 h, at which time the ph of the medium was 7.2 and 25 pg NOT-N ml-l had been produced. The initial ammonium concentration in liquid medium was approximately 5 yg NHZ-N ml-l less than in the absence of montmorillonite because of adsorption by the clay mineral. Growth in the presence of vermiculite was also biphasic (Fig. 1 c). The specific growth rates during the first and second phases of growth were.5 h-l and.21 h-l, respectively. These rates were significantly different from each other (P =.1) but, unlike montmorillonite, vermiculite did not affect the specific growth rate during the first phase (P =.27). The second phase began at h, again when nitrite concentration was approximately 2pg NOT-N ml-l. When either vermiculite or montmorillonite was present, the change in the specific growth rate was not associated with significant change in the ph of the growth medium. Vermiculite also reduced the initial soluble ammonium concentration in the medium but had a lower buffering capacity than montmorillonite. The ph of the medium decreased to 6.5, below the minimum for growth of suspended cells, and the final nitrite concentration was 11-2 pg ml-l. Only one phase of growth was observed in the presence of illite (Fig. 1 d) and the specific growth rate,.48 h-', was not significantly different from that of the control (P =.47). Illite increased the length of the lag phase but did not adsorb ammonium and had little buffering effect. The ph of the medium decreased to 6. but, as with vermiculite, growth continued at ph values below those at which growth occurred in the control culture producing 25 pg NOT-N ml-l. Growth on ammonia-treated vermiculite Following addition of ATV to uninoculated ammoniafree medium, the ammonium concentration increased within 5 h to 2 pg ml-l and then remained constant. In inoculated cultures, the initial increase in ammonium concentration was followed by a reduction to.75 pg NHZ-N ml-l after 7 h, at which time the nitrite concentration was 2.16 pg NO,-N ml-l and a second growth phase began (Fig. 2). The specific growth rates during the first and second phases were.22 h-l, and -9 h-l, which were significantly different from each other (P <-1) and lower than the control (P <O.Ol). The final nitrite concentration was 4.5 yg NOT-N ml-l but the ph did not fall below 7.8. In the absence of clay minerals production of this concentration of nitrite was accompanied by a significant reduction in ph. Containment of ATV within visking tubing reduced the initial ammonium concentration in the medium of uninoculated cultures to.9 pg ml-l (Fig. ). Following

4 1926 S. J. Powell and J. I. Prosser r- I I I 1 I r, I z. z en $ z 6 Fi 2. Changes in nitrite concentration (O), ammonium concentration () and ph () during growth of N. europaea in 15 ml inorganic medium containing ATV. For nitrite concentration, straight lines indicate data used to calculate the specific growth rates (see legend to Fig. 1). 5. M e' c b) -.5 Y.- L Y.- z, - c I 1 I I I I 1 I I 2 z M c.5.z 8 ; 4 Fig.. Changes in nitrite () and ammonium () concentrations during growth of N. europaea in 15 ml inorganic medium containing ATV enclosed within visking tubing and changes in ammonium Concentration (1[) in uninoculated flasks. For nitrite concentrations, straight lines indicate data used to calculate the specific growth rates (see legend to Fig. 1). inoculation with N. europaea, ammonia was utilized for growth and 1.4 pg NOT-N ml-' was produced by 2 h. The specific growth rate during the first 5 h was.1 h-l, significantly lower than in the absence of visking tubing (P < -Ol), and subsequently decreased. The specific growth rate in the second phase was again less than that in the absence of visking tubing (P <-1), but visking tubing alone had no effect on the specific growth rate. Influence of clay minerals on inhibition by nitrapyrin In the absence of clay minerals, addition of nitrapyrin (.5 pg ml-l, final concentration) to exponentially growing cultures reduced the specific growth rate immediately from.5 h-l to.17 h-' (Table 1). The effects of nitrapyrin on the specific growth rate in the presence of clay minerals are also summarized in Table 1. The presence of illite had little effect, but vermiculite and Table 1. Specific growth rates of N. europaea in liquid batch culture in the presence and absence of-5 pg nitrapyrin (NP) ml- and the presence and absence of montmorillonite, vermiculite and illite. ~ ~~~ Specific growth rate (h-i) First phase Second phase Clay type -NP +NP - NP + NP None.5*.17* ND ND Mon tmorilloni te *1*.7* *121* *11* Vermiculite Illite *45*.21 * ND ND ATV Montmorillonite homo-ionic to A1+.29*.26* ND ND ATV homo-ionic to ~ ND, Not determined because growth was not biphasic. * Mean values representing those treatments in which nitrapyrin affected the specific growth rate at the 5% level of significance.

5 Inhibition of clay-colonizing nitrijiers 1927 ATV gave complete protection from inhibition, with no significant effects on the specific growth rate in the first or second phases of growth. In the presence of montmorillonite, nitrapyrin appeared to stimulate the first phase of growth, but reduced the specific growth rate of the second phase to 8% of the control value. This, however, compared with a reduction to 28% in the absence of clay minerals, and montmorillonite therefore also provided protection. Nitrapyrin shortened the length of the first growth phase in the presence of both vermiculite and ATV. Addition of 1 pg nitrapyrin ml-i halted nitrite production in control cultures but in the presence of ATV this concentration did not affect growth in the first phase, and growth in the second phase continued at a reduced rate. ATV homo-ionic to aluminium did not affect the specific growth rate during the first phase but reduced that in the second phase to -22 h-' compared to.55 h-l with untreated ATV. Addition of.5 pg nitrapyrin ml-1 had no effect on nitrite production in the presence of treated ATV. Growth in the presence of montmorillonite treated with aluminium was essentially monophasic. The specific growth rate was similar to that in the absence of clays until 1 h, after which there were slight but insignificant changes in nitrite concentration. Addition of -5 pg nitrapyrin ml-' to exponentially growing cultures containing untreated montmorillonite reduced only the specific growth rate in the second phase. In the presence of montmorillonite homo-ionic to aluminium, addition of nitrapyrin at h immediately reduced the specific growth rate from.29 h-i to.26 h-l (Table 1). Therefore, montmorillonite homoionic to aluminium increased inhibition by nitrapyrin. Hydrolysis o j nitrapyrin The detection limit for the nitrapyrin assay was 2 yg ml-1 and the rate of hydrolysis could not therefore be determined directly at treatment concentrations of.5 and 1. yg ml-i. At higher concentrations, the first-order rate constants for hydrolysis were.21 and.28 h-' in the absence and presence of ATV, respectively, equivalent to half lives of 4 and 25 h, respectively. These differences were not statistically different and there was no evidence of significant adsorption of nitrapyrin by ATV. Discussion SW medium is poorly buffered to facilitate the detection of growth of ammonia oxidizers by acid production, and growth of the strain of N. europaea used in this study ceased as the ph fell to 6.8 when 8 pg NOF-N ml-l had been produced. In the presence of montmorillonite, vermiculite and illite, nitrite yields were greater for equivalent reductions in the ph of the medium and the ph did not change during growth on ATV. Armstrong & Prosser (1988) found that nitrification of ammonia on ATV added to ammonia-free SW medium did not significantly reduce the ph, but when ATV was added to complete medium the ph fell below 7. The presence of ATV increased nitrite yield and also reduced the specific growth rate. Their data are consistent with adsorption and growth of Nitrosornonas on the surface of clay particles with oxidation of ammonia following desorption from the surface. Localized exchange of NH,+ and H+ at the clay surface then effectively immobilizes H+, reducing ph toxicity and acidification of the medium. Reduction in the specific growth rate may be due to lower levels of available ammonia close to the clay surface as a consequence of microbial activity. Underhill & Prosser (1987) demonstrated growth of both ammonia and nitrite oxidizers on the surface of ion-exchange resin beads, with colonization greatest on the surfaces possessing a charge opposite to that of the substrate. These findings imply that in the presence of negatively charged clay minerals, growth and nitrification will occur preferentially at the clay surface. All four clay minerals investigated here have a 2 : 1 lattice structure, representing the ratio of silicon oxide tetrahedra to aluminium oxide (or hydroxide) octahedra as the basic structure. However, their cation exchange capacities (CCs) differ. Montmorillonite, vermiculite and ATV have CC values of 1.24,.95 and.89 meq g-l respectively, while that of illite is.21 meq g-i. For these clays, the CC is correlated with the ability of the mineral layers to expand in water, this ability being limited in illite. These two properties are not, however, directly related and their effects are not separated in this study. xpansion may increase diffusion of ammonium, particularly from ATV, and potentially provide a greater surface area for colonization. Such effects are, however, unlikely to be significant and the effects observed are more likely to result from the CC properties of the clay minerals. The CC of both montmorillonite and vermiculite is indicated by adsorption of ammonium from liquid medium, and both led to biphasic growth. The first phase is believed to result mainly from growth of reversibly bound or suspended cells and the specific growth rates are little affected. During the second phase, a significant proportion of the population is attached irreversibly to the clay surface and their subsequent growth occurs at a lower rate, as on ATV (Armstrong & Prosser, 1988). Attachment and subsequent colonization are not affected by increased ammonium concentration, as ammonium is saturating for growth in the liquid medium, but

6 1928 S. J. Powell and J. I. Prosser are favoured by reduction in ph toxicity. In some cases semilogarithmic plots of nitrite concentration against time for the second phase showed variability, and the slopes, although useful for comparative purposes, are unlikely to be directly equivalent to the specific growth rate. It was not possible to determine from these studies whether reductions in the specific growth rate resulted from substrate limitation or factors specifically associated with surface attachment. Nitrite yield was greatest during growth on montmorillonite, but evidence for localized buffering on vermiculite, similar to that on ATV, was provided by increased nitrite yield and nitrite formation at ph values down to 6.. On these clays, therefore, a reduced specific growth rate was compensated for by increased buffering and greater nitrite oxidation. As a result of its lower expandibility and CC, illite did not decrease the ammonium concentration of SW by adsorption. This reduced its suitability as a site for nitrification monophasic growth was presumably due to growth of suspended, rather than attached cells and the specific growth rate was not reduced. There is evidence for some surface growth as nitrification did proceed at ph values slightly below that possible for suspended cells but the proportion of adsorbed cells was not sufficient to give biphasic growth. These data are in broad agreement with the findings of Kunc & Stotzky (198) and Macura & Stotzky (198) that montmorillonite stimulated soil nitrification but that kaolinite, which also has low CC, had no effect. Direct quantitative comparisons are not possible because of the use of different experimental systems, yielding different kinetic constants, and their use of higher clay concentrations. In addition, the particle size ranges of clay minerals used in this study were large in comparison to those found in the soil. This does not alter conclusions regarding the effects of CC properties but additional effects may arise from colloidal properties of smaller clay particles. Clays with high CC also differed from illite in their effect on inhibition by nitrapyrin, both vermiculite and montmorillonite giving significant protection, while illite had no effect. This may be due to greater susceptibility of free cells. Colonization of montmorillonite and vermiculite is believed to have commenced immediately after inoculation but a significant, irreversibly bound population did not develop until the onset of the second growth phase. Irreversible adsorption of nitrifiers is accompanied by formation of extracellular polymeric material (Cox et al., 198; Keen & Prosser, 1987), which is more extensive in later stages of colonization, but sufficient material may have been present in the second phase to inactivate or immobilize nitrapyrin and reduce its effect. Powell (1985) has shown that established biofilms of Nitrosomonas on glass slides are similarly protected from inhibition, this protection nullifying any possible increase in inhibition through adsorption of nitrapyrin. There was less protection during the first phase, when cells were either freely suspended or reversibly adsorbed. In fact, although data for growth on montmorillonite and vermiculite, following addition of nitrapyrin, were analysed on the basis of biphasic growth, the two phases were less distinct than in non-inhibited cultures. Protection from inhibition may therefore increase continuously as the surface-attached population develops, producing increasing amounts of extracellular material. An alternative, or additional factor is variation in sensitivity with the specific growth rate. xponentially growing cells are more susceptible to inhibition than stationary-phase cells and at low specific growth rates in chemostat culture, nitrapyrin can stimulate nitrite production (Powell, 1985). ATV did not significantly increase the rate of hydrolysis of nitrapyrin and protection was not, therefore, due to a reduced effective concentration of the in hi bi tor. Reduced inhibition at low specific growth rate is most clearly seen during growth on ATV, where specific growth rate is limited by the concentration of ammonium diffusing from the clay particles. This concentration is reduced by enclosing ATV particles within visking tubing, further decreasing the specific growth rate but also significantly lowering the final nitrite concentration. Under these conditions, ammonia was oxidized by freely suspended cells only, indicating that surface growth can increase nitrite yield. Growth on ATV provided complete protection from inhibition by nitrapyrin at a concentration of.5 yg mg-l and inhibition was only partial at a concentration of 1. pg ml-l, which completely inhibited growth of suspended cells. Further evidence for the protective effect of surface growth is provided by experiments in which clays were made homo-ionic to A1+. Although this can potentially influence several aspects of clay mineral structure, the major effect in montmorillonite will be in reducing adsorption of NHZ. This effect will be reduced in ATV, where ammonia is bound more strongly. Thus, growth on montmorillonite treated with A1+ was equivalent to that of freely suspended cells, occurring as a single phase, and cells were not protected from inhibition. In fact, treated montmorillonite enhanced the inhibitory effect of nitrapyrin. Similar treatment of ATV had little effect on growth and no effect on protection from inhibition. In conclusion, the study identifies two major roles for clay minerals in soil nitrification. The first is the ability of expanding clays with high CC, such as montmorillonite and vermiculite, to provide localized exchange of NH,+ and H+ produced during ammonia oxidation, thereby reducing acidification and preventing growth

7 Inhibition of clay-colonizing nitr8ers I929 limitation at low ph. The second is the protection, afforded by surface growth, from inhibition of ammonia oxidation by nitrapyrin. S. J. P. acknowledges receipt of a NRC Postgraduate Research Studentship and we are indebted to Dr J. Russell for provision of clay minerals. References AHRLICHS, G. I., FRASR, A. R. & RUSSLL, J. D. (1972). Interaction of ammonia with vermiculite. Clay Minerals 9, ARMSTRONG,. F. & PROSSR, J. I. (1988). Growth of Nitrosomonas europaea on ammonia-treated vermiculite. Soil Biology and Biochemistry 2, BOZIAN, R. H. & STOTZKY, G. (1976). Inhibition of nitrification in soil by SOz, and the effect of kaolinite and montmorillonite on inhibition. Agronomy Abstracts p. 15. BRIGGS, G. G. (1975). The behaviour of the nitrification inhibitor N- Serve in broadcast and incorporated applications to soil. Journal of the Science of Food and Agriculture 26, BURNS, R. G. (198). Microbial adhesion to soil surfaces: consequences for growth and enzyme activities. In Microbial Adhesion to Surfaces, pp dited by R. C. W. Berkeley, J. M. Lynch, J. Melling, P. R. Rutter & B. Vincent. Chichester: llis Horwood. Cox, D. J., BAZIN, M. J. & GULL, K. (198). Distribution of bacteria in a continuous-flow nitrification column. Soil Biology and Biochemistry 12, GORING, C. A. I. (1962). Control of nitrification by 2-chloro-6- (trichloromethyl) pyridine. Soil Science 9, KN, G. A. & PROSSR, J. I. (1987). The interrelationship between ph and surface growth of Nitrobacter. Soil Biology and Biochemistry 19, KNY, D. R. (1986). Inhibition of nitrification in soils. In Nitrification, pp dited by J. I. Prosser. Oxford: IRL Press. KUNC, F. & STOTZKY, G. (198). Acceleration by montmorillonite of nitrification in soil. Folia Microbiologica 25, MACURA, J. & STOTZKY, G. (198). ffect of montmorillonite and kaolinite on nitrification in soil. Folia Microbiologica 25, POWLL, S. J. (1985). Inhibition of Nitrosomonas europaea by nitrapyrin : the role of surjaces. PhD thesis, University of Aberdeen POWLL, S. J. (1986). Laboratory studies of inhibition of nitrification. In Nitrification, pp dited by J. I. Prosser. Oxford: IRL Press. POWLL, S. J. & PROSSR, J. I. (1986). Inhibition of ammonium oxidation by nitrapyrin in soil and liquid culture. Applied and nvironmental Microbiology 52, RATNAYAK, M. & AUDUS, L. J. (1978). Studies on the effects of herbicides on soil nitrification, 2. Pesticide Biochemistry and Physiology 8, ROWRS, G. A. & ASHWORTH, J. (1982). Bacteriostatic action of nitrification inhibitors. Canadian Journal of Microbiology 28, RUSSLL, J. D. & FRASR, A. R. (1977). Ammonia-treated vermiculite - a possible controlled-release nitrogen fertilizer. Journal of the Science of Food and Agriculture 28, SANTORO, T. & STOTZKY, G. (1967). ffect of electrolyte composition and ph on the particle size distribution of microorganisms and clay minerals as determined by the electrical sensing zone method. Archives of Biochemistry and Biophysics 112, SIMS, R. C. & LITTL, L. (197). nhanced nitrification by addition of clinoptilite to tertiary activated sludge units. nvironmental Letters 4, SKINNR, F. A. & WALKR, N. (1961). Growth of Nitrosomonas europaea in batch and continuous culture. Archiv fur Mikrobiologie 8, UNDRHILL, S.. & PROSSR, J. I. (1985). Inhibition of nitrification by potassium ethyl xanthate in soil and liquid culture. Soil Biology and Biochemistry 17, UNDRHILL, S.. & PROSSR, J. I. (1987). Surface attachment of nitrifying bacteria and their inhibition by potassium ethyl xanthate. Microbial cology 14,