Small undisturbed soil volumes (c. 1·7 cm3) were collected from the surface of a small field plot. Soil volumes were treated with clover-derived substrate, dried and rewetted, or retained continuously moist from the field. These soil volumes were then incubated for 20 days at a matric water potential of either –10 or –30 kPa. At the end of the incubation the soil was analysed for volumetric water content (θv), NO-3 -N, NH+4 -N, total N (%N), and percentages of sand, silt, and clay. The texture terms were included in linear regression models, together with %N and θv as predictors of N mineralisation and nitrification. Clay and sand were often observed to have a significant influence on N mineralisation and nitrification, but silt rarely appeared to influence these processes. In soils retained continuously moist, %clay had a negative relationship with N mineralisation and nitrification, but this relationship was positive in soils that had been dried and rewetted. The results suggest that during periods of relatively high moisture content, soils that are higher in clay are able to protect organic N more effectively from microbial attack. However, on drying and rewetting, the protective mechanisms of clay are undermined, the relatively large protected reservoirs of organic N in high clay soils become more vulnerable to microbial attack, and these soils therefore experience a greater flush of N mineralisation than soils with lower clay levels. The negative influence of clay in the continuously moist soils was not as clearly observed in the soils incubated at –10 kPa as in soils incubated at –30 kPa, suggesting that the decomposition of organic N resident in larger pores (10–30 µm neck diameter) may not be as strongly regulated by clay as that resident in smaller pores. When soils were treated with clover-derived substrate, clay had a positive relationship with N mineralisation and nitrification rates. This may have been because clay limited the diffusion of partially decomposed organics away from the decomposing microbial population, thereby helping to facilitate more complete decomposition of the organic material. Texture had very little influence on the nitrification of urea-derived ammonium.

The existence of microsites of low pH around active colonies of nitrifying soil bacteria has previously been suggested but has been difficult to verify. A study was undertaken to examine whether observed decreases in bulk soil pH that occur during nitrification are in accordance with the theory of acidified nitrification microsites. A red earth soil (sieved <2 mm) was retained at a pH of 5·3 or amended with KHCO3 to achieve a pH of 6·3. Ammonium [(NH4)2SO4] was added to the soils and they were incubated for 35 days. In both soils the pH dropped rapidly and severely limited further nitrification. The soil with the higher initial pH experienced limitations to nitrification at a pH which was 0·2 units higher than that of the soil with the lower initial pH. The explanation for this result is in terms of acidified nitrification microsites. It is suggested that an active nitrifying colony may lower the pH within its immediate vicinity to a critical pH at which nitrification almost ceases. This critical pH achieved at the nitrification microsite is probably unrelated to the initial pH of the soil, but the pH of the soil matrix which is distant from the immediate influence of the nitrification microsite would remain at a pH closer to that of the soil initially. This less acidified region of the soil matrix would have an overriding influence on the measured pH of the bulk soil and account for the discrepancy between the measured pH of the two soils at the end of the incubation. These data provide further evidence that acidified nitrification microsites exist in soil, and that the measured soil pH is a poor estimate of the pH experienced by the microbial biomass.

Increasing laboratory costs and time constraints have stimulated development of techniques which allow the simultaneous extraction of many elements from soil. We assessed several techniques using cation/anion exchange resins which allow the simultaneous extraction of Ca, Mg, K, Al, Mn and P from soil. Elements are extracted by shaking soil either with resin beads or resin membranes in distilled water. The resin is separated from the soil and elements are desorbed using BaCl2/HCl. Concentrations of elements in solution are then determined by conventional chemical methods. Using 50 soils which varied in acidity/fertility, a comparison was made between concentrations of elements extracted by the resin methods and commonly used extraction procedures. Apart from Al, concentrations of elements extracted by the resin procedures correlated well with conventional extraction procedures, although P was less well correlated (R2 < 0.8) than cations (R2 > 0.8). All the resin procedures have the potential to reduce the time required for analysis of Ca, Mg, K, Mn and P in soils. The resin membrane method offers considerable speed and cost advantages over the unconfined resin bead method.

Large numbers of small undisturbed soil volumes (1·7 cm3 ) were collected from the surface layer of a 2 m by 3 m field plot on a red earth near Wagga Wagga, New South Wales. The hypothesis tested was that an analysis of relationships between the volume of different pore size classes and various soil properties, measured on these soil volumes, can provide an understanding of soil organisation within the framework of the pore system. Three discrete findings were presented in confirmation of the hypothesis. (1) A non-uniform distribution of organic N through the pore system was indicated by the data analysis. Soil organic N tended to be concentrated in pores <0·6 µm and in pores 10-30 µm, but not in the intermediate pore size class (pores 0·6-10 µm). Concentrations of organic N in pores <0·6 µm are probably because of physical protection from microbial decomposition, but concentrations of organic N in pores 10-30 µm are probably because these pores are infrequently water-filled, and this limits bacterial activity more severely than in the pores 0·6-10 µm. Currently available assays for potentially mineralisable N cannot account for the effect of substrate location within the pore system, and a characterisation of the soil for the distribution of N in pores may enhance their utility. Soil disturbance is likely to alter organic matter distribution through the pore system and alter mineralisation dynamics. (2) Observations of pore size distributions before and after wetting suggested that soils which were high in organic matter and clay tended to have a greater volume of pores 0 ·6-30 µm which are unstable to drying and rewetting. It is proposed that these unstable pores 0 ·6-30 µm had been produced by the movement and alignment of clay particles during the growth of microbial colonies. (3) The volume of pores <0·6 µm had a relatively strong negative correlation with pH and a relatively strong positive correlation with Mn2+ . A mechanism based on redox chemistry principles was proposed to explain these relationships. It was suggested that the volume of pores <0·6 µm is related to the potential anaerobicity of the soil volume. In anaerobic conditions, the terminal electron acceptor for organic C oxidation may be MnO2 instead of O2, and in these circumstances, considerably more H+ would be consumed than in aerobic conditions. It is suggested that this alkaline effect extends into regions of the matrix where N mineralisation and nitrification are occurring, and stimulates these processes. The greater nitrification which may result from such a chain of events may, over time, effect greater acidification in those soil volumes which have greater microporosity.

Natural heterogeneity of soil properties was used to explore their influence on nitrogen (N) mineralisation and nitrification in undisturbed small soil volumes (soil cells; c. 1 · 7 cm3 ) sampled from a small field plot (2 m by 3 m). Soil cells (840) were randomly ascribed to 1 of 6 treatments in which soils were retained continuously moist (M10 and M30 treatments) and amended with organic N from clover (Cl10 and Cl30 treatments), dried and rewetted (DW10), or treated with urea (Ur10) (subscripts indicate soil incubation at matric potential - 10 or - 30 kPa). After 20 days of incubation at 24C, each soil cell was analysed for NO-3 -N, NH + 4 -N, pH, bulk density (BD), volumetric water content (θv), water content at - 490 kPa (θv490), and pH buffer capacity (pHBC). On 25 soil cells from each treatment, % clay, % silt, % sand, total N (% N), organic carbon (% C), and 7 cations and anions were also determined. Net N mineralisation and net nitrification occurred in all treatments, and the total mineral N at the end of the incubation was 497, 81, 73, 31, 27, and 31 µg N/g in the Ur10 Cl10, Cl30, M10, M30, and DW10 treatments, respectively. Net N mineralisation in the M30 treatment was 84% of that in the M10 treatment, and net N mineralisation in the Cl30 treatment was 86% of that in the Cl10 treatment. Fluctuations in soil pH varied markedly between treatments and over time, and it was apparent that alkaline processes were occurring in all soil cells. The heterogeneity between soil samples was substantial for all of the soil variables. Soil variables were classified in a hierarchy from the least to the most fundamental based on their stability through time. This ranking provides a conceptual tool for understanding interrelationships between soil properties and for interpreting results of regression analyses. The sampling approach adopted in this study was designed to harness the natural heterogeneity of soil properties in the small field site while keeping other properties and environmental factors, that usually vary over larger distances, constant. Both the extent of heterogeneity of soil properties and the nature of their correlations with NO-3 -N suggested that this technique would be useful in the exploration of how soil properties influence N mineralisation and nitrification.