挺好的

1006A. K. Patra et al.

increases in denitri cation and nitri cation activities havebeen observed in grassland ecosystems in response tolong-term grazing pressure (Parsons et al., 1991; Groff-man et al., 1993; Frank and Groffman, 1998; Frank et al.,2000; Le Roux et al., 2003; Patra et al., 2005), but studieson the effect of grazing on free-living N2- xation are limited(Patra et al., 2005). Similarly, effects of mowing on N uxes and N retention in grasslands have been reported(Maron and Jeffries, 2001). Grazing and mowing can alsoaffect the size and composition of key microbial functionalgroups driving N dynamics, although most of publishedstudies have not quanti ed the in uence of grazing/mow-ing per se without the confounding effect of inorganic Nfertilization (Clegg et al., 1998; Webster et al., 2002; butsee Patra et al., 2005).

Changes in the activity and genetic structure of thesesoil bacterial functional groups induced by managementthrough grazing/mowing can be explained by differentdirect and indirect effects, including (i) input of urine anddung by herbivores (Ruess and McNaughton, 1987), (ii)changes in soil porosity through trampling of soil by ani-mals (Abdelmagid et al., 1987), (iii) changes in competi-tion for N between microorganisms and the recentlydefoliated sward (Busso et al., 2001) and (iv) changes inthe input rates and quality of plant residues and rootexudates due to defoliation (Paterson and Sim, 1999;Hamilton and Frank, 2001). However, grazing/mowingregime also strongly modi es the identity of major plantspecies (Collins et al., 1998; Olff and Ritchie, 1998), andit has been shown that plant species can in uence nitri -cation, denitri cation and N2- xation enzyme activities(Wheatley et al., 1990; Crush, 1998; Priha et al., 1999;Van der Krift and Berendse, 2001; Briones et al., 2002).Similarly, plant species can in uence the structure of soilmicrobial communities (Ibekwe and Kennedy, 1998; Prihaet al., 1999; Briones et al., 2002; Söderberg et al., 2002).Thus, the response of nitrifying, denitrifying and N2- xingcommunities to long-term grazing/mowing pressureobserved at the ecosystem level could be also indirectlydriven or in uenced by differences in the identity in majorplant species. To our knowledge, this hypothesis has beentested once and only for nitri cation and denitri cationenzyme activities (Le Roux et al., 2003).

The objectives of this study were (i) to unravel the directeffect of management regime and the effect of changes

in the identity major plant species on the enzyme activityand genetic structure of bacterial functional groupsinvolved in soil N dynamics (nitri ers, denitri ers and free-living N2- xers); those groups have different ecologicalrequirements and it was expected that their response tomanagement and plant species could differ, (ii) to test ifchanges in soil moisture and/or mineral N could explainthe observed effects on community activity and structureand (iii) to test if changes in activity were correlated or notto changes in genetic structure. This was achieved bycomparing enzyme activities and genetic structures ofthese functional groups below three dominant grass spe-cies, whose abundance was favoured, discouraged orunaffected by management through grazing/mowing,between intensively managed (I) or extensively managed(E) grassland sites. Soil moisture and nitrate and ammo-nium concentrations were measured on the same sam-ples. The substrate-induced respiration (SIR) and thegenetic structure of the eubacterial community were alsomeasured to test whether the observed changes in thebacterial functional groups involved in soil N dynamicsoccurred in a background of changes in microbial biomassand in the dominant soil bacterial populations.Results

Management and plant species effects on soil moisture and mineral N concentrations

The effect of plant species on soil nitrate concentrationwas signi cant, whereas the effects of managementregime and management–plant interaction were not sig-ni cant (Table 1, Fig. 1). Nitrate concentration values werealways higher than 19 µg N g 1 (Fig. 1). Nitrite concentra-tion was always below detection limit. The effect of man-agement on soil ammonium concentration was signi cant,whereas the effects of plant species and management–plant interaction were not signi cant (Table 1, Fig. 1).Ammonium values were low, particularly under plants onE plots (< 0.12 µg N g 1, as compared with values around0.3 µg N g 1 under plants on I plots). Soil moisture tendedto be slightly higher under I than E treatments (Table 1,Fig. 1), but differences between I and E treatments werenot signi cant for all species (Fig. 1). Soil moisture washigher under Holcus lanatus than other plant species,particularly for E conditions (Fig. 1).

Table 1.Results (P-values) of the two-way ANOVA for the effects of management regime, plant species, and management–species interaction onnitrate and ammonium concentrations, soil moisture, and enzyme activities of soil microbial communities.

Nitrate

Management regimePlant species

Management–species

NS0.028NS

Ammonium0.0001NSNS

Moisture0.0170.01NS

SIRNSNSNS

Free-living N2- xation0.00380.0290.0057

Denitri cationNS0.037NS

Nitri cation0.0001NS0.024

NS, not signi cantly different at level P = 0.05.

© 2006 The Authors

Journal compilation © 2006 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 8, 1005–1016