Herbivores can indirectly affect ecosystem productivity by modifying feedbacks that occur between dominant plants and below-ground properties, especially through altering nutrient availability in soil. The aim of this study was to examine, under controlled conditions, the effect of simulated browsing by large herbivores on the growth characteristics of downy birch (Betula pubescens), a dominant tree species of native regenerating forests in northern Britain, and to determine how effects of browsing on tree growth cascaded through to soil microbial communities, thereby affecting nutrient availability in soil. Downy birch seedlings were grown in mesocosms for 2 years and subjected to simulated browsing in the form of defoliation and clipping treatments. Upon destructive harvest, a number of measures of both tree growth characteristics and soil biological and nutrient properties were made. Clipping of birch trees significantly reduced total root biomass (27%), fine root biomass (29%), coarse root biomass (27%) and above-ground biomass (18%), whereas defoliated trees were significantly shorter than non-defoliated trees. Despite these significant and negative effects of browsing on tree growth, soil biological properties remained largely unaffected, other than rates of N mineralisation, which were greater under defoliated trees. We conclude that other factors, such as herbivore effects on litter quantity and quality which feedback to soil biological properties in the longer-term are more important in determining ecosystem responses to browsing.

Our aim was to determine if soil ergosterol concentration provides a quantitative estimate of the soil fungal biomass concentration, as is usually assumed. This was done by comparing soil ergosterol measurements with soil fungal biomass (fungal biomass C) concentrations estimated by microscopic measurements and by the selective inhibition technique linked to substrate-induced respiration (SIR). The measurements were compared in a silty-clay loam soil given a range of previous treatments designed to increase or decrease the soil fungal biomass and so also to change the soil ergosterol concentration. The treatments used were ryegrass amendment, to increase the total and fungal biomass, and CHCl3-fumigation and the addition of the biocides, captan, bronopol and dinoseb, to decrease both ergosterol and fungal biomass C concentrations. The mineralization of ergosterol following addition to sand innoculated with soil extract, and to a sandy loam soil, was also determined. The added ergosterol was little, if at all, degraded following addition to either sand or the unfumigated or fumigated soil during a 10 d aerobic incubation. Similarly, pesticide addition did not significantly change soil ergosterol concentrations yet the soil fungal biomass C concentration decreased significantly. Thus, the ratio: (soil ergosterol concentration/soil fungal biomass C concentration) was much higher in the pesticide-treated soils than the control soil. Following ryegrass amendment, soil ergosterol concentration increased from about 6-12 microgram(-1) soil within 5 d and then decreased gradually to about 7 microgram g(-1) soil by 20 d incubation. Changes in fungal biomass C (measured by direct microscopy) closely mirrored changes in soil ergosterol over this period. However, when the amended soil was fumigated and then incubated for a further 5 d, the initial ergosterol concentration declined from 7 to 5 microgram g(-1) soil by 20 d incubation (a decline of about 0.4). The comparable decline in fungal biomass C was about eight-fold. Thus the ratio of ergosterol to fungal biomass C increased from 0.005 to about 0.01. There was a significant correlation (r>0.84, P<0.001) between soil ergosterol concentration and fungal biomass measured by either SIR or microscopy. However, three data points played a vital role in the correlation. When these points were excluded the relationship was very poor (r<0.4). Our results therefore suggest that substantial amounts of ergosterol may exist, other than in living cells, for considerable periods, with little, if any mineralization. Thus, these results indicate that ergosterol and fungal biomass C concentrations are not always closely correlated, due to the slow metabolism of ergosterol in recently dead fugal biomass and/or the existence of exocellular ergosterol in soil.

During the first few days after rewetting of an air-dried soil (AD-RW), microbial activity increases compared to that in the original moist soil, causing increased mineralisation (a flush) of soil organic carbon (C) and other nutrients. The AD-RW flush is believed to be derived from the enhanced mineralisation of both non-biomass soil organic matter (due to its physical release and enhanced availability) and microbial biomass killed during drying and rewetting. Our aim was to determine the effects of AD-RW on the mineralisation of soil organic matter and microbial biomass during and after repeated AD-RW cycles and to quantify their proportions in the CO2-C flushes that resulted. To do this, a UK grassland soil was amended with 14C-labelled glucose to label the biomass and then given five AD-RW cycles, each followed by 7 d incubation at 25 degrees C and 50% water holding capacity. Each AD-RW cycle increased the amount of CO2-C evolved (varying from 83 to 240 microgram g(-1) soil), compared to the control with, overall, less CO2-C being evolved as the number of AD-RW cycles increased. In the first cycle, the amount of biomass C decreased by 44% and microbial ATP by 70% while concentrations of extractable C nearly doubled. However, all rapidly recovered and within 1.3 d after rewetting, biomass C was 87% and ATP was 78% of the initial concentrations measured prior to air-drying. Similarly, by 2 d, extractable organic C had decreased to a similar concentration to the original. After the five AD-RW cycles, the amounts of total and 14C-labelled biomass C remaining in the soil accounted for 60 and 40% of those in the similarly incubated control soil, respectively. Soil biomass ATP concentrations following the first AD-RW cycle remained remarkably constant (ranging from about 10 to 14 micromol ATP g(-1) biomass C) and very similar to the concentration in the fresh soil prior to air-drying. We developed a simple mathematical procedure to estimate the proportion of CO2-C derived from biomass C and non-biomass C during AD-RW. From it, we estimate that, over the five AD-RW cycles, about 60% of the CO2-C evolved came from mineralisation of non-biomass organic C and the remainder from the biomass C itself.

The Ni-hyperaccumulating plant Alyssum murale accumulates exceptionally high concentrations of nickel in its aboveground biomass. The reasons for hyperaccumulation remain unproven; however, it has been proposed that elemental allelopathy might be important. High-Ni leaves shed by the plant may create a "toxic zone" around the plant where germination or growth of competing plants is inhibited. The efficacy of this argument will partially depend upon the rate at which leaves degrade in soil and free metals are released, and the subsequent rate at which metals are bound to soil constituents. To test the degradation of biomass of hyperaccumulators, A. murale was grown on both high- and low-Ni soils to achieve high- (12.0 g Ni/kg) and low- (0.445 g Ni/kg) Ni biomass. Shredded leaf and stem biomass were added to a serpentine soil from Oregon that was originally used to grow high-Ni biomass and a low-Ni control soil from Maryland. Biomass Ni was readily soluble and extractable, suggesting near immediate release as biomass was added to soil. Extractable nickel in soil amended with biomass declined rapidly over time due to Ni binding in soil. These results suggest that Ni released from biomass of Ni hyperaccumulators may significantly affect their immediate niche only for short periods of time soon after leaf fall, but repeated application may create high Ni levels under and around hyperaccumulators.

We investigated the relationship between soil nutrients and fine-root biomass at broad (among ecosystem types) and fine (within a 20 m x 20 m plot) spatial scales in forested wetlands of the southeastern United States. We selected three replicates each of high-fertility floodplain swamps, low-fertility depressional swamps, and intermediate-fertility river swamp sloughs and measured soil nutrient availability (NO3-N, NH4-N, and PO4-P) and fine-root biomass. At one replicate of each wetland type, a dense network of sampling points was used to measure variability (variance and coefficient of variation) of soil nutrients and fine-root biomass. At the broad scale, fine-root biomass was lower in floodplain swamps than in either river swamp sloughs or depressional swamps. Also, multiple linear regression and Spearman's rank correlations indicated a negative relationship between soil nutrient availability and fine-root biomass. Fine-scale correlates between soil nutrient availability and fine-root biomass were generally weak. Fine-scale variability of NO3-N and NH4-N was greatest in the floodplain swamps, but nutrients were not spatially patchy at any of the sampled sites. We conclude that soil nutrient availability may control fine-root biomass at the broad scale, but it is unclear if the same is true at fine spatial scales.

Variations in dynamics of shoot and root systems might be linked to differential genotypic capacities to withstand soil moisture deficit. Two pot experiments were conducted to evaluate the root and shoot dynamics of three maize genotypes under two soil moisture regimes. Improved Tiniguib, an open pollinated variety (OPV) and C-818, a single-cross hybrid were used in experiment 1 while G-404 (or Ghen 404), a three-way cross, and C-818 were used in the second (or follow-up) experiment. The open pollinated variety had significantly higher shoot biomass under drought conditions while the single-cross hybrid C-818 had similar shoot biomass under drought and well-watered conditions. Genotype x water regime interaction effect on shoot dry weight involving the two hybrids was not significant indicating that both hybrids had similar shoot biomass dynamics under drought and well-watered conditions. The three genotypes had the largest bulk of their root biomass in the 0-15 cm layer of the soil. Improved Tiniguib had lower root biomass than C-818 in the 0-15, 15-30, 30-45, 45-60 and 60-75 layers of the soil throughout the 20-day soil drying period. G-404 had significantly higher root biomass in the 30-45 cm layer at 21 days after withholding water (DAWW) but had significantly lower root biomass in the 45-60 cm layer at 28 DAWW than C-818. Root biomass was not significantly related to root length and soil moisture content. Hence, root biomass is not a reliable indicator of root absorbing capacity.

The relationships among soil microbial biomass, pH and available of heavy metal fractions were evaluated in longterm contaminated soils during an incubation experiment with the amendment of zeolite (natural clinoptilolite) and the subsequent addition of glucose. The values of pH after the addition of glucose decreased during the first day of incubation approximately at about one unit and corresponded with the maximum increase of microbial biomass. The available heavy metal contents extracted by H2O, 1 mol/l NH4NO3 and 0.005 mol/l DTPA increased during the first two days of incubation. Only a few significant relationships were found between the available metal contents and pH or microbial biomass. This fact could be ascribed to the different dynamics of the microbial biomass, pH and metal availability after glucose addition, when the highest metal contents during the incubation were usually reached one day later in respect to the greatest changes of pH and microbial activity. In comparison to soils without zeolite addition, the variants with natural clinoptilolite showed lower heavy metal contents in all used extractants with the exception of Cd which in H2O extracts tended to increase.

Land use practices exert a major influence on plant productivity, soil and plant nutrient content, and within-stand nutrient cycling in wetlands in agricultural landscapes. We examined differences between improved and seminative pastures in plant and soil nutrient characteristics in seasonally flooded wetlands in subtropical grazing land of south central Florida. The wetlands were embedded within either grazed improved pastures with a long-term history of fertilizer application or seminative pastures with no history of previous fertilizer application. Soil nutrient concentrations decreased with soil depth for both land use types. Total C, N, and P were significantly greater (P < 0.05) in the 0- to 15-cm mineral layer compared with the deeper layers (15–30, 30–45 cm) for both improved and seminative pasture wetland soils. Improved pasture wetlands had greater amounts of total P (22.3 kg P ha⁻¹) in the upper 0- to 15-cm soil layer than did the seminative pasture wetlands (15.7 kg P ha⁻¹). Plant and soil (0–15 cm) N/P and C/P ratios were lower in improved pasture wetlands compared with seminative pasture wetlands, suggesting greater P enrichment in improved pasture wetlands. Microbial biomass C and N decreased with soil depth in both pasture types. Soil microbial biomass C/total C ratios decreased with soil depth and were similar for both improved and seminative pasture wetlands. Our results suggest that plant and soil nutrient enrichment and storage in temporary wetlands may be impacted by adjacent land use practices, which potentially leads to the alteration of the structure and functions of these wetland ecosystems.

The contribution of cacti and shrubs to root biomass and fine-root production was described in a semiarid Mexican scrub. Both life forms were evaluated for fine-root production variation in relation to changes of nitrogen in the soil, with a fertilization experiment. Cacti represented 78% of the total mean root biomass (660 +/- 70 g.m(-2) (mean +/- SE)) in the complete soil profile (50 cm in depth). In both life forms, root biomass was higher near the surface of the soil. Roots <3 mm in diameter represented 92.5% for cactus root biomass and 69.4% for shrubs. Monthly root biomass varied significantly between months, and significant differences were obtained between plant life forms. Fine and very fine root production was estimated as 3.76 Mg.ha(-1).year(-1), and cactus contribution to total root production was 81.2%. Significant differences were obtained between life forms. It was clear that a low concentration in the soil nitrogen diminishes fine-root production, supporting the hypothesis that in arid ecosystems nitrogen is a limiting factor for primary production.