Do interactions with soil organisms mediate grass responses to defoliation?

ARTICLE IN PRESS

Soil Biology & Biochemistry 40 (2008) 894–905 www.elsevier.com/locate/soilbio

Do interactions with soil organisms mediate grass responses to defoliation? Katja Ilmarinena,, Juha Mikolaa,1, Mauritz Vestbergb a

Department of Biological and Environmental Science, University of Jyva¨skyla¨, P.O. Box 35, FI-40014 Jyva¨skyla¨, Finland b Plant Production Research, MTT Agrifood Research Finland, Antinniementie 1, FI-41330 Vihtavuori, Finland Received 11 May 2007; received in revised form 5 November 2007; accepted 6 November 2007 Available online 3 December 2007

Abstract Defoliation-induced changes in grass growth and C allocation are known to affect soil organisms, but how much these effects in turn mediate grass responses to defoliation is not fully understood. Here, we present results from a microcosm study that assessed the role of arbuscular mycorrhizal (AM) fungi and soil decomposers in the response of a common forage grass, Phleum pratense L., to defoliation at two nutrient availabilities (added inorganic nutrients or no added nutrients). We measured the growth and C and N allocations of P. pratense plants as well as the abundance of soil organisms in the plant rhizosphere 5 and 19 d after defoliation. To examine whether defoliation affected the availability of organic N to plants, we added 15N-labelled root litter to the soil and tracked the movement of mineralized 15N from the litter to the plant shoots. When inorganic nutrients were not added, defoliation reduced P. pratense growth and root C allocation, but increased the shoot N concentration, shoot N yield (amount of N in clipped plus harvested shoot mass) and relative shoot N allocation. Defoliation also reduced N uptake from the litter but did not affect total plant N uptake. Among soil organisms, defoliation reduced the root colonization rates of AM fungi but did not affect soil microbial respiration or the abundance of microbe-grazing nematodes. These results indicate that interactions with soil organisms were not responsible for the increased shoot N concentration and shoot N yield of defoliated P. pratense plants. Instead, these effects apparently reflect a higher efficiency in N uptake per unit plant mass and increased relative allocation of N to shoots in defoliated plants. The role of soil organisms did not change when additional nutrients were available at the moment of defoliation, but the effects of defoliation on shoot N concentration and yield became negative, apparently due to the reduced ability of defoliated plants to compete for the pulse of inorganic nutrients added at the moment of defoliation. Our results show that the typical grass responses to defoliation—increased shoot N concentration and shoot N yield—are not necessarily mediated by soil organisms. We also found that these responses followed defoliation even when the ability of plants to utilize N from organic sources, such as plant litter, was diminished, because defoliated plants showed higher N-uptake efficiency per unit plant mass and allocated relatively more N to shoots than non-defoliated plants. r 2007 Elsevier Ltd. All rights reserved. Keywords: AM fungi; Clipping; Decomposer food web; Grassland; Phleum pratense; Plant carbon allocation; Plant nitrogen dynamics; Plant–soil interactions; Soil nematodes; 15N

1. Introduction It is increasingly recognized that soil organisms may play an important role in the response of plants to defoliation Corresponding author.

E-mail addresses: [email protected].fi (K. Ilmarinen), juha.mikola@helsinki.fi (J. Mikola), mauritz.vestberg@mtt.fi (M. Vestberg). 1 Present address: Department of Ecological and Environmental Sciences, University of Helsinki, Niemenkatu 73, FI-15140 Lahti, Finland. 0038-0717/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.soilbio.2007.11.004

(Bardgett and Wardle, 2003; Wardle et al., 2004). For instance, defoliation typically increases the N concentration of grass shoots (Wilsey et al., 1997; Fahnestock and Detling, 1999; Green and Detling, 2000), suggesting that this is a consequence of improved availability of N in the soil following defoliation (Holland and Detling, 1990). For example, Hamilton and Frank (2001) recently showed how defoliation of smooth meadow-grass Poa pratensis L. increased the availability of N in the soil and led to elevated shoot N concentrations in defoliated plants. In

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

addition, they found that the increased availability of N in the soil was associated with increased growth of decomposer microbes in the plant rhizosphere, indicating that the elevated shoot N concentrations were a result of defoliation-induced soil microbial activity. However, increased shoot N concentrations may also reflect other changes occurring after defoliation. Arbuscular mycorrhizal (AM) fungi, in addition to enhancing plant P uptake, also substantially improve uptake of N (Ma¨der et al., 2000). The majority of studies (reviewed by Gehring and Whitham, 1994) have reported decreases in AM colonization rates of plant roots after defoliation, but increases have also been recorded (Hartley and Amos, 1999; Hokka et al., 2004; Kula et al., 2005), which implies that elevated shoot N concentrations may in some cases also reflect enhanced AM colonization rates. Elevated shoot N concentrations can also reflect greater relative allocation of nutrients to shoots after defoliation (Ruess, 1988; Louahlia et al., 2000), and in principle they could simply result from a greater reduction in plant C assimilation than N acquisition during recovery from defoliation. There is therefore a need for studies that provide a comprehensive assessment of the role of different soil organisms, in relation to other mechanisms, in plant responses to defoliation. The potential role of soil organisms in the response of plants to defoliation stems from the fact that defoliation induces changes in plant C allocation that can affect the availability of labile C in the soil. For instance, defoliation can increase photosynthate allocation to roots (Holland et al., 1996) or increase the concentrations of soluble C in roots (Paterson and Sim, 1999), which in turn may increase C exudation from roots to soil and stimulate soil microbes (Mawdsley and Bardgett, 1997). Similarly, increased allocation of C to roots after defoliation may induce greater AM infection rates of roots (Gehring and Whitham, 1994). However, a stimulated soil microbial community does not necessarily improve soil nutrient supply for plants. Soil microbes compete effectively with plants for available nutrients (Van Veen et al., 1989; Kaye and Hart, 1997; Bardgett et al., 2003), and C addition and the resulting microbial growth may, in contrast, reduce plant nutrient uptake (Diaz et al., 1993; Jonasson et al., 1996; Schmidt et al., 1997). Microbe-feeding animals, such as protists and nematodes, that release nutrients from microbial biomass when consuming it are therefore necessary for maintaining the availability of nutrients for plants (Clarholm, 1985; Paterson, 2003). When microbial grazers are present, increased root C release can potentially improve plant N availability if it induces microbes to increase utilization of dead organic matter, because part of the organic N assimilated into the microbial biomass at this stage will subsequently be released by microbial grazers for plant uptake. To understand how trophic interactions in soil are affected by defoliation and what consequences this may have on N mineralization are therefore crucial to assessing the role of soil organisms in plant responses to

895

defoliation, such as elevated shoot N concentrations. Although evidence has long been accumulating of soil decomposers being affected by plant defoliation (Stanton, 1983; Seastedt et al., 1988; Mawdsley and Bardgett, 1997; Fu et al., 2001; Mikola et al., 2001, 2005a; Techau et al., 2004), no study has so far experimentally linked the multitrophic level effects of defoliation in soil food webs to soil N availability and plant N uptake. Here, we present a greenhouse study that was aimed at assessing the role of soil organisms in grass responses to defoliation. We established controlled microcosms consisting of natural soil communities and seedlings of timothy Phleum pratense L., an important forage grass in northern pastures (Berg et al., 1996). We defoliated the seedlings and followed their growth and C and N allocations as well as the abundances of soil organisms in the plant rhizosphere. We also added 15N-labelled litter in the soil before defoliation and tracked the movement of 15N from litter to plant shoots to examine whether defoliation affected the availability of soil organic N to plants. We predict that if interactions with soil organisms mediate P. pratense responses to defoliation and especially the expected increase in shoot N concentrations, (1) the activity and abundance of soil decomposers and/or rates of root AM colonization will increase after defoliation, and as a result (2) transfer of N from the added litter to P. pratense plants will be higher for defoliated than for non-defoliated plants. Since we reasoned that the availability of easily utilizable inorganic forms of N in the soil at the moment of defoliation may determine whether defoliation induces microbes to increase utilization of N from the litter added, we further measured the effects of defoliation in systems with and without additional inorganic nutrients. This experimental approach allowed us to compare a comprehensive and previously unexplored set of mechanisms potentially able to mediate and modify grass responses to defoliation. 2. Materials and methods 2.1. Establishment of the microcosms Soil with a sandy texture, total C content of 2.5%, total N content of 1.5 g kg1, total P content of 1.2 g kg1 and pH (water) of 6.4 was collected from a pasture in central Finland (621N, 261E) in September 2003. The soil was sieved (1 cm) and homogenized, and 950 g (dry weight equivalent) was added to each of 68 plant pots (height 9 cm, diameter 11–13 cm, with holes in the base to allow free drainage of water). No organisms were removed from or added to the soil during this procedure. To allow later addition of 15N-labelled litter into the soil, three plastic tubes (diameter 1.6 cm, height 13 cm) were placed vertically into the full depth of the soil in a triangle in the middle of each pot, 3.5 cm apart. One 5-wk-old P. pratense seedling, sprouted in vermiculite, was then planted in each pot in the middle of the triangle.

ARTICLE IN PRESS 896

K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

Four weeks after planting, 15N-labelled P. pratense root litter (total N concentration, 0.84% of dry weight; 15N 12.35 at%) was added to each microcosm. The litter was produced in a sand culture of P. pratense seedlings, using an 15NH415NO3-enriched nutrient solution prepared according to Ingestad (1979). The roots were washed free of sand, dried and cut into 1-cm-long fragments. The three plastic tubes were then removed from the soil and each hole was filled with a mixture of soil (21.7 g dry weight per hole) and 15N-labelled litter (0.27 g dry weight per hole). The locations of the three litter patches were marked for later recognition. For measuring the background 15N concentrations of plant material in different combinations of treatments, eight additional microcosms were established in January 2005 without litter amendment. 2.2. Growth conditions The microcosms were kept in a greenhouse on a plastic tray. The soil was kept moist by frequently filling the tray with tap water for 1–2 h, and to equalize the environmental conditions for the microcosms their placements on the tray were rearranged twice per week. The sunlight was supplemented with eight 400-W daylight lamps for 14 h d1. The photosynthetic photon flux density varied between 117 and 203 mmol m2 s1 at the height of the plant shoots, the relative atmospheric humidity varied between 40% and 70%, and the temperature was 18–20 1C during the light period and 14 1C at night. 2.3. Experimental design The experimental design involved three treatment factors—defoliation, nutrient addition and harvest time— in a fully factorial design, with eight replicates of each treatment combination (64 experimental pots in total). Defoliation consisted of two treatment categories, i.e. no trimming vs. trimming of P. pratense seedlings to 8 cm above the soil surface 7 d after litter addition, which removed about 60% of the shoot biomass (recommended height of cattle grazing for P. pratense is 8–10 cm above the soil surface; Virkaja¨rvi et al., 2001). Prior to the defoliation treatment, four surplus microcosms were harvested to ensure that the roots had proliferated in the litter patches by the time of defoliation. Nutrient addition had two treatment categories, i.e. no added nutrients vs.12 ml (containing 62 mg N; corresponding to about 50 kg N ha1) nutrient solution, prepared according to Ingestad (1979), added to the soil simultaneously with the defoliation treatment. To ensure a uniform spread of nutrients in the soil, 2 ml of solution were injected into each of the three litter patches and 2 ml into three spots between the patches. In microcosms receiving no additional nutrients, 2 ml water was added to the corresponding six spots. Harvest time also consisted of two treatment categories, i.e. harvests 5 d vs.19 d after the defoliation. Previous experiments showed that this time scale is appropriate for detecting the response

of soil organisms to plant defoliation (Mikola et al., 2005a). In the eight additional microcosms established for exploring the background 15N concentrations, defoliation and nutrient addition treatments (two replicates for each treatment combination) were applied when the seedlings were 7 wk old. The shoot material was collected 5 d after defoliation. 2.4. Sampling procedure At each harvest, the shoot biomass was first removed, dried (70 1C, 48 h) and weighed. The three litter patches were then removed, using a soil corer, and pooled to yield a single litter patch sample. To collect rhizosphere soil, the plant roots were gently shaken. The soil that remained attached to the roots after shaking was considered as rhizosphere soil and collected for further analyses. The roots were then washed over a sieve with tap water, dried (70 1C, 48 h) and weighed. The concentrations of total N, 15 N and C in the harvested shoot and root mass were analysed at Iso Analytical Ltd., Sandbach, Cheshire, UK. The root 15N data were, however, not used because some root samples were apparently contaminated with pieces of labelled litter. The excess 15N in the shoot mass (calculated by subtracting the background values from the measured values) was used to calculate the amount of total N transferred from the litter to the plant shoots. To evaluate the AM fungal colonization of roots, a 0.7-g subsample of fresh roots was preserved in 50% alcohol. The samples were later bleached in 10% KOH overnight, acidified in 1% HCl and stained at 90 1C for 1 h with 0.01% methyl blue (Phillips and Hayman, 1970; Grace and Stribley, 1991). The mycorrhizal structures were investigated under a high-power microscope using 30 root fragments, each 1 cm long, from each stained sample. The method of Trouvelot et al. (1986), including the ‘Mycocalc’ program downloaded from http://www.dijon.inra.fr/mychintec, was then used to estimate the overall AM root colonization rate (% of colonized root fragments in total number of fragments investigated), the colonization intensity of colonized roots (index of how densely the colonized fragments were colonized) and the arbuscular abundance of colonized roots (index of how densely the colonized fragments were packed with arbuscules). To evaluate the effects of defoliation and litter addition on soil decomposer organisms, soil microbial respiration and the abundance of microbe-feeding nematodes were measured separately for both rhizosphere and litter patch soil. Microbial respiration was measured within 24 h after each harvest, using 3.5–4.0 g (fresh mass) subsamples of soil. All living root material was first removed from the soil using forceps. The samples were then placed in sealable glass bottles, incubated for 4 h at 22 1C and respiration determined as the total CO2–C released between 1 and 4 h of incubation, using infrared gas analysis (Universal Carbon Analyser EQ92). After incubation, the bottles were dried (70 1C, 24 h) to determine the dry mass of the

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

897

soil samples used. The nematodes were extracted from 15-g (fresh mass) subsamples of soil, using the wet funnel method (Sohlenius, 1979). The total number of nematodes was first counted live and later, using preserved samples, all individuals (or 150 nematodes per sample if the total number exceeded this) were identified to genus and allocated to trophic groups according to Yeates et al. (1993).

(Quinn and Keough, 2003). To compare microbial respiration and the abundance of nematode trophic groups between the litter patch and rhizosphere soils, a t-test for paired samples was used.

2.5. Statistical analyses

3.1. Plant growth

The effects of defoliation, nutrient addition and harvest time on plant attributes and soil biota were analysed using a three-way analysis of variance (ANOVA) (carried out with SPSS 12.0; SPSS Inc., Chicago, IL, USA). In the case of litter patch microbial respiration, the soil water content was included in the ANOVA model as a covariate. The homogeneity of variances was tested, using Levene’s test, and when needed the data were transformed to meet the assumptions of the ANOVA. In the case of predatory nematodes, however, no transformation was effective and the analyses were conducted with heterogeneous variances. If the ANOVA resulted in significant (Po0.05) interactions among defoliation, nutrient addition and harvest, the factors were fixed individually and the effects of other factors were analysed within the levels of the fixed factor

Defoliation reduced shoot production (comprising both trimmed and harvested shoot material) and harvested shoot mass at both harvests (Table 1), while nutrient addition increased shoot production (F ¼ 218, Po0.001) and harvested shoot biomass (F ¼ 238, Po0.001) at the second harvest only (Fig. 1a and b). The root biomass was reduced by defoliation at both harvests (Table 1), reduced by nutrient addition at the first harvest (F ¼ 4.6, Po0.05) and increased by nutrient addition at the second harvest (F ¼ 63, Po0.001) (Fig. 1c). Defoliation reduced the proportion of shoot mass in the total harvested plant mass at the first harvest (F ¼ 97, Po0.001), but showed no effect at the second harvest (F ¼ 3.0, P40.05), whereas nutrient addition increased the shoot proportion at both harvests (Table 1, Fig. 1d).

3. Results

Table 1 F-statistics from three-way ANOVA (n ¼ 8) of the effects of defoliation (no trimming vs. single trimming), nutrient addition (no added nutrients vs. addition of nutrient solution) and harvest time (5 d vs. 19 d after defoliation and nutrient addition treatments) on Phleum pratense plants Dependent variable

Source of variation Defoliation, F1,56

Nutrient addition, F1,56

Time, F1,56

D  N, F1,56

D  Time, F1,56

N  Time, F1,56

D  N  Time, F1,56

Plant growth Shoot production Harvested shoot mass Harvested root mass Proportion of shoot mass to harvested plant mass

137 474 103 57.0

105 149 16.7 32.6

1199 1487 471 91.6

0.77 0.73 0.92 2.22

7.81 13.7 3.95 27.0

118 92.8 50.4 0.30

0.15 o0.01 0.50 0.56

Plant C allocation Shoot C concentration Root C concentration Ratio of root C content to plant C yield

2.60 0.08 4.77

0.58 4.32 19.5

38.9 160 14.4

0.59 0.05 5.85

0.02 0.25 0.02

2.92 0.05 0.82

1.69 0.29 0.35

12.6 1.64 26.8 86.0 29.0

459 112 554 580 81.0

1001 120 132 287 29.0

51.7 24.1 82.4 67.3 29.2

1.54 96.1 4.12 6.76 0.22

178 79.5 27.6 29.1 2.96

3.93 9.16 0.24 0.82 0.16

10.4

112

109

16.5

4.98

60.5

3.28

Plant N allocation and uptake Shoot N concentration Root N concentration Shoot N yield Plant N yield Ratio of shoot N yield to plant N yield Litter-derived N in shoot production

Shoot production consists of defoliated and harvested shoot mass, shoot N yield includes the N content of defoliated and harvested shoot mass, and plant C and N yield include the C and N content of defoliated and harvested shoots and harvested roots.  Po0.05. Po0.01. Po0.001.

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

898

HARVEST 1

2.5 2.0 1.5 1.0 0.5 0.0 N

C

1.0 0.8 0.6 0.4 0.2 0.0 C

N

C

HARVEST 2

2.5 2.0 1.5 1.0 0.5 0.0

N

Proportion of shoot mass to harvested plant mass

C

Harvested roo tmass (g dry weight)

HARVEST 2 Harvested shoot mass (g dry weight)

Total shoot production (g dry weight)

HARVEST 1

C

N

C

N

C

N

C

N

1.0 0.8 0.6 0.4 0.2 0.0

N

Fig. 1. Plant growth (mean+1 S.D.; n ¼ 8) in Phleum pratense microcosms in relation to defoliation (no trimming vs. single trimming), nutrient addition (no added nutrients vs. addition of nutrient solution at the moment of defoliation) and harvest time (5 d vs. 19 d after defoliation and nutrient addition treatments): (a) total shoot production (consists of defoliated and harvested shoot mass), (b) harvested shoot mass, (c) harvested root mass and (d) proportion of shoot mass to harvested plant mass. White and hatched bars represent non-defoliated and defoliated systems, respectively. N denotes nutrient addition, C denotes control for nutrient addition, and harvests 1 and 2 represent the two harvests performed 5 and 19 d after defoliation and nutrient addition treatments, respectively.

HARVEST 2

HARVEST 1 Root C concentration (% of dry weight)

Shoot C concentration (% of dry weight)

HARVEST 1 46 44 42 40 38

HARVEST 2

46 44 42 40 38 36

36 N

C

Ratio of root C content to plant C yield

C

N

N

C

C

N

0.5 0.4 0.3 0.2 0.1 0.0 C

N

C

N

Fig. 2. Plant C allocation (mean+1 S.D.; n ¼ 8) in Phleum pratense microcosms in relation to defoliation (no trimming vs. single trimming), nutrient addition (no added nutrients vs. addition of nutrient solution at the moment of defoliation) and harvest time (5 d vs. 19 d after defoliation and nutrient addition treatments): (a) shoot C concentration, (b) root C concentration and (c) relative root C allocation, or ratio of root C content to plant C yield (plant C yield consists of C content of defoliated and harvested shoot mass and C content of harvested roots). Treatment symbols are as in Fig. 1.

3.2. Plant C allocation Defoliation reduced the relative root C allocation (i.e. the ratio of the harvested root C content to total plant C

yield) in microcosms where nutrients were not added (F ¼ 15, Po0.001), but did not affect the shoot and root C concentrations (Table 1, Fig. 2a–c). Nutrient addition decreased the root C concentration (Table 1) and relative

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

root C allocation when the plants were not defoliated (F ¼ 33, Po0.001), but did not affect the shoot C concentration (Table 1, Fig. 2a–c). 3.3. Plant N uptake and allocation In microcosms where nutrients were not added, defoliation increased the shoot N concentration (F ¼ 58, Po0.001) and shoot N yield (i.e. the total N content of trimmed and harvested shoot material) (F ¼ 7.7, Po0.05) by an average of 28% and 12%, respectively, but showed no effect on the total plant N yield (i.e. the sum of the shoot N yield and N content of the harvested root material) (F ¼ 0.60, P40.05) (Fig. 3a, c and d). When the nutrients were added, defoliation decreased the shoot N concentration by 8% at the first harvest (F ¼ 12, Po0.001), and shoot N yield (F ¼ 103, Po0.001) and total plant N yield (F ¼ 155, Po0.001) at both harvests by an average of 36% (Fig. 3a, c and d). The root N concentration was reduced by defoliation by 28% at the first harvest when nutrients were added (F ¼ 86, Po0.001), but was promoted by defoliation

3.4. Arbuscular mycorrhizal fungi The overall AM colonization rate of plant roots was reduced by defoliation at both harvests (Table 2). Within

HARVEST 2

4 3 2 1

HARVEST 1

0 C

N

C

50 40 30 20 10

4 3 2 1 0 C

N

C

N

C

N

C

N

C

N

50 40 30 20 10 0

0 N

C

N

Litter derived N in shoot production (mg)

C

Ratio of shoot N yield to plant N yield

HARVEST 2

5

N

Plant N yield (mg)

Shoot N yield (mg)

by an average of 16% at the second harvest at both nutrient addition levels (F ¼ 35, Po0.001) (Fig. 3b). Nutrient addition increased the shoot N concentration, shoot N yield and total plant N yield at each harvest (Table 1), whereas the root N concentration was increased at the first harvest only (F ¼ 182, Po0.001) (Fig. 3a–d). The relative shoot N allocation (i.e. the ratio of the shoot N yield to total plant N yield) was increased by defoliation when nutrients were not added (F ¼ 78, Po0.001), but not when nutrients were added (Fo0.01, P40.05) (Fig. 3e). The amount of litter-derived N in shoot production was not affected by defoliation when nutrients were not added (F ¼ 0.50, P40.05), but was decreased by 23% when nutrients were added (F ¼ 30, Po0.001) (Fig. 3f). Nutrient addition increased the relative shoot N allocation and the amount of litter-derived N in shoot production (Table 1) (Fig. 3e and f).

Root N concentration (% of dry weight)

Shoot N concentration (% of dry weight)

HARVEST 1 5

899

0.8 0.6 0.4 0.2 0.0 C

N

C

N

0.4 0.3 0.2 0.1 0.0 C

N

Fig. 3. Plant N allocation and uptake (mean+1 S.D.; n ¼ 8) in Phleum pratense microcosms in relation to defoliation (no trimming vs. single trimming), nutrient addition (no added nutrients vs. addition of nutrient solution at the moment of defoliation) and harvest time (5 d vs. 19 d after defoliation and nutrient addition treatments): (a) shoot N concentration, (b) root N concentration, (c) shoot N yield (consists of N content of defoliated and harvested shoot mass), (d) plant N yield (consists of shoot N yield and N content of harvested roots, (e) relative shoot N allocation, or ratio of shoot N yield to plant N yield and (f) litter-derived N found in shoot production. Treatment symbols are as in Fig. 1.

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

900

Table 2 F-statistics from three-way ANOVA (n ¼ 6–8) of the effects of defoliation (no trimming vs. single trimming), nutrient addition (no added nutrients vs. addition of nutrient solution) and harvest time (5 d vs. 19 d after defoliation and nutrient addition treatments) on soil organisms in Phleum pratense microcosms Dependent variable

AM colonization of plant roots AM colonization rate Colonization intensity of colonized roots Arbuscular abundance in colonized roots Soil microbial respiration Basal respiration in rhizosphere soila Basal respiration in litter patch soilb

Source of variation Defoliation, F1,56

Nutrient addition, F1,56

Time, F1,56

6.24 1.81

1.67 17.7

27.8 2.04

0.01 0.24

0.21 6.82*

8.77 o0.01

1.32 0.63

4.01

19.6

28.9

0.17

4.81

3.00

1.35

0.36

0.12

10.2

0.07

0.68

1.97

0.84

2.07

0.91

18.0

1.03

0.13

4.86*

1.85

3.87 12.7 1.88 1.39 0.26 1.92 0.95 0.03 1.50 0.24

5.34* 78.7 0.69 0.32 0.74 76.8 1.5 0.28 0.01 19.4

0.94 2.32 0.04 1.42 0.78 o0.01 1.07 0.05 0.10 2.22

0.34 0.29 0.43 o0.01 2.61 3.32 o0.01 0.06 1.50 5.48

1.35 4.56 0.62 4.17 0.18 1.15 0.43 1.92 0.38 0.05

2.40 1.90 0.03 0.32 0.62 11.7 0.96 0.33 o0.01 0.87

Abundance of nematode trophic groups Bacterivores in rhizosphere soil 0.04 Bacterivores in litter patch soil 1.51 Fungivores in rhizosphere soil 0.30 Fungivores in litter patch soil 2.10 Omnivores in rhizosphere soil 2.93 Omnivores in litter patch soil 3.93 Herbivores in rhizosphere soil 0.61 Herbivores in litter patch soil 0.21 Predators in rhizosphere soil 0.39 Predators in litter patch soil 5.78

D  N, F1,56

D  Time, F1,56

N  Time, F1,56

D  N  Time, F1,56

a

Error d.f. ¼ 53. Error d.f. ¼ 55.  Po0.05.  Po0.01.  Po0.001. b

the colonized root parts, the colonization intensity (F ¼ 7.8, Po0.01) and arbuscular abundance (F ¼ 8.8, Po0.01) were, however, reduced at the first harvest only (Fig. 4a–c). Nutrient addition decreased the overall root colonization rate of AM fungi at the second harvest (F ¼ 9.0, Po0.01), but had negative effects on colonization intensity and arbuscular abundance at both harvests (Table 2, Fig. 4a–c). 3.5. Soil microbes and nematodes Soil microbial respiration was not affected by defoliation or nutrient addition in the rhizosphere or litter patch soil (Table 2, Fig. 5a). However, respiration was greater in the litter patch than rhizosphere soil at the first harvest (t ¼ 3.7, P ¼ 0.001, n ¼ 29) (Fig. 5a). Herbivorous and bacterivorous nematodes were the most abundant nematode trophic groups in the rhizosphere soil, constituting 48% and 42% of all nematode individuals, respectively. However, bacterivores comprised 80% of all individuals in the litter patch soil (Fig. 5b–f). Defoliation and nutrient addition did not affect the abundance of nematode trophic groups in the rhizosphere soil (Table 2), whereas in the litter patch soil nutrient

addition increased the numbers of bacterivores (F ¼ 16, Po0.001) and fungivores (F ¼ 5.2, Po0.005) at the first harvest (Fig. 5b–f). In the litter patch soil, defoliation increased the abundance of omnivores at the first harvest but only in systems that did not receive additional nutrients (F ¼ 13, Po0.001) (Fig. 5d). At the first harvest, nematode abundances were greater in the litter patch than rhizosphere soil for bacterivores (t ¼ 7.3, Po0.001, n ¼ 32), fungivores (t ¼ 3.4, P ¼ 0.002), omnivores (t ¼ 5.6, Po0.001) and predators (t ¼ 3.5, P ¼ 0.001), while the abundance of herbivores was greater in the rhizosphere soil (t ¼ 11, Po0.001) (Fig. 5b–f). At the second harvest, the litter patch soil still supported greater abundances than did the rhizosphere soil for bacterivores (t ¼ 3.0, P ¼ 0.006) and fungivores (t ¼ 5.3, Po0.001), but the reverse was true for herbivores (t ¼ 12, Po0.001), omnivores (t ¼ 3.2, P ¼ 0.003) and predators (t ¼ 2.5, P ¼ 0.017) (Fig. 5b–f). 4. Discussion The purpose of our study was to assess the relative role of soil organisms in the response of P. pratense seedlings to

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

HARVEST 1

HARVEST 1

HARVEST 2 Colonization intensity of colonized roots

AM colonization rate (% of root fragments)

80 60 40 20 0 N

C

Arbuscular abundance in colonized roots

C

HARVEST 2

50 40 30 20 10 0 C

N

901

N

C

N

120 100 80 60 40 20 0 C

N

C

N

Fig. 4. Arbuscular mycorrhizal colonization of plant roots (mean+1 S.D.; n ¼ 8) in Phleum pratense microcosms in relation to defoliation (no trimming vs. single trimming), nutrient addition (no added nutrients vs. addition of nutrient solution at the moment of defoliation) and harvest time (5 d vs. 19 d after defoliation and nutrient addition treatments): (a) AM colonization rate (% of colonized root fragments in total number of fragments investigated), (b) colonization intensity of colonized roots (index of how densely the colonized fragments were colonized) and (c) arbuscular abundance of colonized roots (index of how densely the colonized fragments were packed with arbuscules). Treatment symbols are as in Fig. 1.

defoliation. Although our data are correlative and cannot prove causality between the variables measured, the relative roles of various mechanisms covering the response of P. pratense seedlings to defoliation still appear clear in our study. We found that when inorganic nutrients were not added, defoliation reduced P. pratense shoot and root growth and increased the N concentration and N yield of P. pratense shoots. This agrees with earlier studies, reviewed by Ferraro and Oesterheld (2002), which found that defoliation reduces shoot and root growth but increases the shoot N concentrations of grasses (Wilsey et al., 1997; Fahnestock and Detling, 1999; Green and Detling, 2000). However, soil microbial activity (measured as microbial respiration) and the abundance of microbial grazers were not affected and the root colonization rates of AM fungi were retarded by defoliation in our study, thus indicating that soil organisms were not responsible for the elevated shoot N concentrations. This interpretation is further supported by our finding that defoliation did not affect total plant N uptake, but instead, promoted relative shoot N allocation. This means that the elevated shoot N concentrations and shoot N yield of defoliated P. pratense seedlings were, in our study, explained solely by the higher efficiency of N uptake (equal N uptake with reduced plant mass) and increased allocation of N to shoots in defoliated plants. Defoliation reduced the relative C allocation to P. pratense roots in our study (even though the root C concentration was not altered), along with decreased overall colonization rate, colonization intensity and arbus-

cular abundance of AM fungi in P. pratense roots. Mycorrhizal fungi are known to consume a substantial proportion of plant photosynthate production (Smith and Read, 1997) and the negative effects of defoliation on AM colonization rates of plant roots are believed to be associated with photosynthate limitation occurring after defoliation (Caldwell et al., 1981; Gehring and Whitham, 1994; Strauss and Agrawal, 1999). The negative effects of defoliation on the mycorrhizal attributes 5 d after defoliation are consistent with this idea. However, we also found that 2 wk later the two attributes that best describe the functional properties of the mycorrhizae, i.e. the colonization intensity and arbuscular abundance, had already recovered from defoliation, despite the overall AM colonization rate of roots still remaining depressed. This supports earlier findings that functionally important structures of AM fungi recover more rapidly from defoliation than does the average root colonization rate (Klironomos et al., 2004). Despite having clear effects on AM fungi, defoliation did not affect the activity of soil microbes or the abundance of microbial grazers in the P. pratense rhizosphere. Since the numbers of microbial grazers are known to closely reflect the variation in root C release (Christensen et al., 1992, 2007), these results indicate that defoliation did not affect the amount of C released from P. pratense roots. Similar results were recently presented by Bazot et al. (2005), who found that although defoliation induced changes in C allocation and root soluble C concentration in field-grown perennial rye-grass Lolium perenne L. plants, it did not

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

902

RHIZOSPHERE SOIL HARV 1 HARV 2

LITTER PATCH SOIL HARV 1 HARV 2 40 Bacterivorous nematodes (no. [g soil d.w.]-1)

Microbial respiration (µg CO2-Ch-1 [g soil d.w.]-1)

4 3 2 1 0 N

C

N

C

N

C

20

10

N

C

N

C

N

N

C

N

C

N

C

C

N

N

C

C

N

C

3 Omnivorous nematodes (no. [g soil d.w.]-1)

3 Fungivorous nematodes (no. [g soil d.w.]-1)

30

0 C

2

1

2

1

0

0 C

N

C

N

C

N

C

N

N

C

N

3 Predatory nematodes (no. [gsoil d.w.]-1)

10 Herbivorous nematodes (no. [g soil d.w.]-1)

LITTER PATCH SOIL HARV 1 HARV 2

RHIZOSPHERE SOIL HARV 1 HARV 2

8 6 4 2 0

2

1

0 C

N

C

N

C

N

C

N

C

N

C

N

Fig. 5. Microbial respiration and abundance of nematode trophic groups in rhizosphere and litter patch soil (mean+1 S.D.; n ¼ 6–8) in Phleum pratense microcosms in relation to defoliation (no trimming vs. single trimming), nutrient addition (no added nutrients vs. addition of nutrient solution at the moment of defoliation) and harvest time (5 d vs. 19 d after defoliation and nutrient addition treatments): (a) microbial respiration, (b) bacterivorous, (c) fungivorous, (d) omnivorous, (e) herbivorous and (f) predatory nematodes. Treatment symbols are as in Fig. 1.

affect C availability or the abundance of bacteria and bacteria-feeding nematodes in the L. perenne rhizosphere. In contrast to defoliation, litter addition had a clear positive effect on microbes and their grazers in our experiment, and since nutrient addition did not have such an effect, microbial growth was apparently C-limited. This means that had root C exudation been significantly altered by defoliation, microbes and their grazers should have responded to it. Defoliation can have negative (Mikola and Kyto¨viita, 2002; Nguyen and Henry, 2002; Dilkes et al., 2004), neutral (Bazot et al., 2005) and positive (Holland et al., 1996; Paterson and Sim, 1999, 2000; Hamilton and Frank, 2001) effects on grass root C release, and similar

contrasting responses to defoliation can be found in microbes (Guitian and Bardgett, 2000; Hamilton and Frank, 2001; Sankaran and Augustine, 2004; Bazot et al., 2005) and microbial grazers (Stanton, 1983; Todd et al., 1992; Todd, 1996; Mikola et al., 2001; Ilmarinen et al., 2005). Combined with our results, this indicates that the role of soil organisms in plant responses to defoliation may be highly context-dependent. Defoliation-induced increases in root C exudation can increase soil microbial biomass, soil N availability and plant N uptake (Hamilton and Frank, 2001). In our systems without added nutrients, defoliation did not affect total plant N uptake and the abundances of microbe-feeding

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

nematodes, which are known to closely follow soil microbial production (Christensen et al., 1992) and enhance mineralization of N assimilated into microbial biomass (Ferris et al., 1998). This agrees with the idea that growth of soil decomposers is a prerequisite for increased plant N uptake after defoliation. The amount of litter-derived N in shoot production was not affected by defoliation, but since defoliation increased the relative shoot N allocation, this actually suggests that the total uptake of N from the litter (shoots combined with roots) was lower in defoliated than in non-defoliated plants. We did not measure soil microbial 15 N content and therefore do not know whether microbial utilization of litter N differed among the defoliation treatments. The reason for the lower litter-N uptake of defoliated plants is therefore not certain, but these findings may reflect the impaired AM structure and reduced root biomass of the defoliated P. pratense plants, which could weaken the ability of the plants to utilize N from the patchily distributed plant litter. In contrast, combination of the results of total plant N yield (not affected by defoliation) and shoot and root production (reduced by defoliation) indicates that plant uptake efficiency of other N sources (amount of N taken up per unit shoot and root mass) was higher for defoliated than non-defoliated seedlings. This supports the hypothesis that elevated shoot N concentrations could simply be a consequence of a greater reduction in C assimilation relative to N acquisition following defoliation. Defoliation also increased the root N concentration at the later harvest, which supports the idea that aboveground herbivory enhances root quality by increasing root N concentrations (Seastedt et al., 1988). This indicates that defoliation may have longer-term effects on soil organisms and N mineralization through enhanced decomposition of root litter in such cases as ours, where no shortterm effects were observed. We speculated that if easily assimilated inorganic forms of N are abundant in the rhizosphere when a plant is defoliated, activation of microbes and their grazers by a C pulse may not lead to enhanced plant N uptake from the added litter, because the microbes would then mostly utilize the inorganic forms of nutrients. Since microbes and their grazers were not activated by defoliation in our study, we could not properly test this hypothesis. However, it appears that the increased availability of inorganic nutrients in the soil affected the response of P. pratense seedlings to defoliation. For instance, defoliation reduced shoot and plant N yield, root and shoot N concentrations and the amount of litter-derived N in shoot production in systems where nutrients were added, which contrasts with the effects of defoliation in systems where additional nutrients were not available. Defoliation is known to reduce root nutrient uptake (Jarvis and Macduff, 1989; Donaghy and Fulkerson, 1998) and the negative effects of defoliation on shoot and root N concentrations and plant N yield are probably explained by the decreased ability of defoliated plants to compete with microbes for the pulse of mineral N that was available at the moment of defoliation.

903

Defoliation led to decreased amounts of litter-derived N in shoot production probably because it did not increase the relative shoot N allocation in fertilized systems, which could have counteracted the generally decreased litter-N uptake. As in N allocation, C allocation was not affected by defoliation when nutrients were added. This may have resulted from the fact that the effects of nutrient addition and defoliation on plant C and N allocation were parallel—both treatments decreased the relative root C allocation and increased the relative shoot N allocation. When plants change their allocation pattern due to fertilization, simultaneously occurring defoliation probably does not impact as heavily, and vice versa. We also found that nutrient addition increased the abundance of microbial-feeding nematodes in the litter patch soil and that this was associated with improved plant N uptake from the litter. This indicates that nutrient addition had such a positive effect on N availability in litter patches that we expected to occur after defoliation as a consequence of increased C deposition. Our results show how the commonly occurring consequences of defoliation among graminoids, such as increased shoot N concentration and shoot N yield, are not necessarily mediated by soil organisms or the availability of nutrients in soil, as earlier suggested by Hamilton and Frank (2001). It appears that, due to the higher efficiency of N uptake per unit plant mass and increased shoot N allocation in defoliated plants, defoliation can increase shoot N yield even when the ability of plants to utilize N from patchily distributed dead organic matter is diminished as the result of reduced root biomass and AM colonization rate of roots. When these results are contrasted with those obtained by Hamilton and Frank (2001), who stressed the role of soil organisms, it appears that factors such as plant species identity and/or growing conditions may significantly affect the role of soil organisms in grass responses to defoliation. This notion agrees well with previous findings that grass species differ in how they allocate resources between shoots and roots after defoliation (Wilsey et al., 1997). Little is known of the response of P. pratense to defoliation, but the available evidence indicates that P. pratense may belong to a group of grasses that does not rely on soil organisms when recovering from defoliation. For instance, it was shown that the release of C assimilates from P. pratense roots is decreased rather than increased after defoliation (Mikola and Kyto¨viita, 2002) and that defoliation has few shortterm (this study) or long-term (Mikola et al., 2005b) effects on soil organisms in the P. pratense rhizosphere. In the field, P. pratense also shows lower regrowth capacity after defoliation than meadow fescue Lolium pratense (Huds.) S.J. Darbyshire (formerly Festuca pratensis Huds.), another important forage grass in northern pastures (Virkaja¨rvi et al., 2001). However, the available evidence is not unequivocal, since Hokka et al. (2004) found that repeated defoliation increased P. pratense root C concentration (despite decreasing shoot C concentration), abundance of

ARTICLE IN PRESS 904

K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905

bacteria-feeding nematodes in the P. pratense rhizosphere, soil NH4–N concentration (although not statistically significant) and P. pratense shoot N concentration. Unlike other results from P. pratense investigations, this pattern matches well with that found by Hamilton and Frank (2001) for P. pratensis, indicating that other contextdependent factors in addition to plant species identity may also affect the degree to which interactions with soil organisms contribute to grass responses to defoliation. Acknowledgements We are grateful to David Wardle and two anonymous reviewers for their helpful and insightful comments on an earlier version of the manuscript, Iuliana Popovici for identification of the nematodes, Roger Jones for advice on the 15N technique, Satu Ala-Ko¨nni, Sohvi Ja¨ntti, Leena Kontiola, Marko Moilanen and Kirsi Ruotsalainen for assistance at the greenhouse and laboratory work and James Thompson for language revision. This study is part of the project ‘Linking belowground food webs, plants and aboveground herbivores in grassland ecosystems’, funded by the Academy of Finland and the University of Jyva¨skyla¨. References Bardgett, R.D., Wardle, D.A., 2003. Herbivore-mediated linkages between aboveground and belowground communities. Ecology 84, 2258–2268. Bardgett, R.D., Streeter, T.C., Bol, R., 2003. Soil microbes compete effectively with plants for organic-nitrogen inputs to temperate grasslands. Ecology 84, 1277–1287. Bazot, S., Mikola, J., Nguyen, C., Robin, C., 2005. Defoliation-induced changes in carbon allocation and root soluble carbon concentration in field-grown Lolium perenne plants: do they affect carbon availability, microbes and animal trophic groups in soil? Functional Ecology 19, 886–896. Berg, C.C., McElroy, A.R., Kunelius, H.T., 1996. Timothy. In: Mosel, L.E., Buxton, D.R., Casler, M.D. (Eds.), Cool Season Forage Grasses. American Society of Agronomy, Madison, WI, pp. 643–664. Caldwell, M.M., Richards, J.H., Johnson, D.A., Nowak, R.S., Dzurec, R.S., 1981. Coping with herbivory: photosynthetic capacity and resource allocation in two semiarid Agropyron bunchgrasses. Oecologia 50, 14–24. Christensen, H., Griffiths, B., Christensen, S., 1992. Bacterial incorporation of tritiated thymidine and populations of bacteriophagous fauna in the rhizosphere of wheat. Soil Biology & Biochemistry 24, 703–709. Christensen, S., Bjørnlund, L., Vesterga˚rd, M., 2007. Decomposer biomass in the rhizosphere to asses rhizodeposition. Oikos 116, 65–74. Clarholm, M., 1985. Interactions of bacteria, protozoa and plants leading to mineralization of soil nitrogen. Soil Biology & Biochemistry 17, 181–187. Diaz, S., Grime, J.P., Harris, J., McPherson, E., 1993. Evidence of a feedback mechanism limiting plant response to elevated carbon dioxide. Nature 364, 616–617. Dilkes, N.B., Jones, D.L., Farrar, J., 2004. Temporal dynamics of carbon partitioning and rhizodeposition in wheat. Plant Physiology 134, 706–715. Donaghy, D.J., Fulkerson, W.J., 1998. Priority for allocation of watersoluble carbohydrate reserves during regrowth of Lolium perenne. Grass and Forage Science 53, 211–218.

Fahnestock, J.T., Detling, J.K., 1999. Plant responses to defoliation and resource supplementation in the Pryor Mountains. Journal of Range Management 52, 263–270. Ferraro, D.O., Oesterheld, M., 2002. Effect of defoliation on grass growth. A quantitative review. Oikos 98, 125–133. Ferris, H., Venette, R.C., van der Meulen, H.R., Lau, S.S., 1998. Nitrogen mineralization by bacterial-feeding nematodes: verification and measurement. Plant and Soil 203, 159–171. Fu, S., Kisselle, K.W., Coleman, D.C., Hendrix, P.F., Crossley Jr., D.A., 2001. Short-term impacts of aboveground herbivory (grasshopper) on the abundance and 14C activity of soil nematodes in conventional tillage and no-till agroecosystems. Soil Biology & Biochemistry 33, 1253–1258. Gehring, C.A., Whitham, T.G., 1994. Interactions between aboveground herbivores and the mycorrhizal mutualists of plants. Trends in Ecology and Evolution 9, 251–255. Grace, C., Stribley, D.P., 1991. A safer procedure for routine staining of vesicular–arbuscular mycorrhizal fungi. Mycological Research 95, 1160–1162. Green, R.A., Detling, J.K., 2000. Defoliation-induced enhancement of total aboveground nitrogen yield of grasses. Oikos 91, 280–284. Guitian, R., Bardgett, R.D., 2000. Plant and soil microbial responses to defoliation in temperate semi-natural grassland. Plant and Soil 220, 271–277. Hamilton, E.W., Frank, D.A., 2001. Can plants stimulate soil microbes and their own nutrient supply? Evidence from a grazing tolerant grass. Ecology 82, 2397–2402. Hartley, S.E., Amos, L., 1999. Competitive interactions between Nardus stricta L. and Calluna vulgaris (L.) Hull: the effect of fertilizer and defoliation on above- and below-ground performance. Journal of Ecology 87, 330–340. Hokka, V., Mikola, J., Vestberg, M., Seta¨la¨, H., 2004. Interactive effects of defoliation and an AM fungus on plants and soil organisms in experimental legume-grass communities. Oikos 106, 73–84. Holland, E.A., Detling, J.K., 1990. Plant response to herbivory and belowground nitrogen cycling. Ecology 71, 1040–1049. Holland, J.N., Cheng, W., Crossley Jr., D.A., 1996. Herbivore-induced changes in plant carbon allocation: assessment of below-ground C fluxes using carbon-14. Oecologia 107, 87–94. Ilmarinen, K., Mikola, J., Nieminen, M., Vestberg, M., 2005. Does plant growth phase determine the response of plants and soil organisms to defoliation? Soil Biology & Biochemistry 37, 433–443. Ingestad, T., 1979. Mineral nutrient requirements of Pinus silvestris and Picea abies seedlings. Physiologica Plantarum 45, 373–380. Jarvis, S.C., Macduff, J.H., 1989. Nitrate nutrition of grasses from steadystate supplies in flowering solution culture following nitrate deprivation and/or defoliation. I. Recovery of uptake and growth and their interactions. Journal of Experimental Botany 40, 965–976. Jonasson, S., Vestergaard, P., Jensen, M., Michelsen, A., 1996. Effects of carbohydrate amendments on nutrient partitioning, plant and microbial performance of a grasslsnd-shrub ecosystem. Oikos 75, 220–226. Kaye, J.P., Hart, S.C., 1997. Competition for nitrogen between plants and soil microorganisms. Trends in Ecology and Evolution 12, 139–143. Klironomos, J.N., McCune, J., Moutoglis, P., 2004. Species of arbuscular mycorrhizal fungi affect mycorrhizal responses to simulated herbivory. Applied Soil Ecology 26, 133–141. Kula, A.A.R., Hartnett, D.C., Wilson, G.W.T., 2005. Effects of mycorrhizal symbiosis on tallgrass prairie plant–herbivore interactions. Ecology Letters 8, 61–69. Louahlia, S., Laine´, P., Thornton, B., Ourry, A., Boucaud, J., 2000. The  role of N-remobilisation and the uptake of NH+ 4 and NO3 by Lolium perenne L. in laminae growth following defoliation under field conditions. Plant and Soil 220, 175–187. Ma¨der, P., Vierheilig, H., Streitwolf-Engel, R., Boller, T., Frey, B., Christie, P., Wiemken, A., 2000. Transport of 15N from a soil compartment separated by a polytetrafluoroethylene membrane to

ARTICLE IN PRESS K. Ilmarinen et al. / Soil Biology & Biochemistry 40 (2008) 894–905 plant roots via the hyphae of arbuscular mycorrhizal fungi. New Phytologist 146, 155–161. Mawdsley, J.L., Bardgett, R.D., 1997. Continuous defoliation of perennial ryegrass (Lolium perenne) and white clover (Trifolium repens) and associated changes in the composition and activity of the microbial population of an upland grassland soil. Biology and Fertility of Soils 24, 52–58. Mikola, J., Kyto¨viita, M.-M., 2002. Defoliation and the availability of currently assimilated carbon in Phleum pratense rhizosphere. Soil Biology & Biochemistry 34, 1869–1874. Mikola, J., Yeates, G.W., Barker, G.M., Wardle, D.A., Bonner, K.I., 2001. Effects of defoliation intensity on soil food-web properties in an experimental grassland community. Oikos 92, 333–343. Mikola, J., Nieminen, M., Ilmarinen, K., Vestberg, M., 2005a. Belowground responses by AM fungi and animal trophic groups to repeated defoliation in an experimental grassland community. Soil Biology & Biochemistry 37, 1630–1639. Mikola, J., Ilmarinen, K., Nieminen, M., Vestberg, M., 2005b. Long-term soil feedback on plant N allocation in defoliated grassland miniecosystems. Soil Biology & Biochemistry 37, 899–904. Nguyen, C., Henry, F., 2002. A carbon-14-glucose assay to compare microbial activity between rhizosphere samples. Biology and Fertility of Soils 35, 270–276. Paterson, E., 2003. Importance of rhizodeposition in the coupling of plant and microbial productivity. European Journal of Soil Science 54, 741–750. Paterson, E., Sim, A., 1999. Rhizodeposition and C-partitioning of Lolium perenne in axenic culture affected by nitrogen supply and defoliation. Plant and Soil 216, 155–164. Paterson, E., Sim, A., 2000. Effect of nitrogen supply and defoliation on loss of organic compounds from roots of Festuca rubra. Journal of Experimental Botany 51, 1449–1457. Phillips, J.M., Hayman, D.S., 1970. Improved procedures for clearing roots and staining parasitic and vesicular–arbuscular mycorrhizal fungi for rapid assessment of infection. British Mycological Society Transactions 55, 158–161. Quinn, G.P., Keough, M.J., 2003. Experimental Design and Data Analysis for Biologists. Cambridge University Press, Cambridge. Ruess, R.W., 1988. The interaction of defoliation and nutrient uptake in Sporobolus kentrophyllus, a short-grass species from the Serengeti Plains. Oecologia 77, 550–556. Sankaran, M., Augustine, D.J., 2004. Large herbivores suppress decomposer abundance in a semiarid grazing ecosystem. Ecology 85, 1052–1061. Schmidt, I.K., Michelsen, A., Jonasson, S., 1997. Effects of labile soil carbon on nutrient partitioning between an arctic graminoid and microbes. Oecologia 112, 557–565.

905

Seastedt, T.R., Ramundo, R.A., Hayes, D.C., 1988. Maximization of densities of soil animals by foliage herbivory: empirical evidence, graphical and conceptual models. Oikos 51, 243–248. Smith, S.E., Read, D.J., 1997. Mycorrhizal Symbiosis. Academic Press, London. Sohlenius, B., 1979. A carbon budget for nematodes, rotifers and tardigrades in a Swedish coniferous forest soil. Holarctic Ecology 2, 30–40. Stanton, N.L., 1983. The effect of clipping and phytophagous nematodes on net primary production of blue grama, Bouteloua gracilis. Oikos 40, 249–257. Strauss, S.Y., Agrawal, A.A., 1999. The ecology and evolution of plant tolerance to herbivory. Trends in Ecology and Evolution 14, 179–185. Techau, M.E., Bjørnlund, L., Christensen, S., 2004. Simulated herbivory effects on rhizosphere organisms in pea (Pisum sativum) depend on phosphate. Plant and Soil 264, 185–194. Todd, T.C., 1996. Effects of management practices on nematode community structure in tallgrass prairie. Applied Soil Ecology 3, 235–246. Todd, T.C., James, S.W., Seastedt, T.R., 1992. Soil invertebrate and plant responses to mowing and carbofuran application in a North American tallgrass prairie. Plant and Soil 144, 117–124. Trouvelot, A., Kough, J.L., Gianinazzi-Pearson, V., 1986. Mesure du taux de mycorrhization VA d’un syste`me radiculaire. Recherche de methods d’estimation ayant une signification fonctionelle. In: GianinazziPearson, V., Gianinazzi, S. (Eds.), Fysiological and Genetical Aspects of Mycorrhizae. INRA Press, Paris, pp. 217–221. Van Veen, J.A., Merckx, R., Van de Geijn, S.C., 1989. Plant- and soil related controls of the flow of carbon from roots through the soil microbial biomass. Plant and Soil 115, 179–188. Virkaja¨rvi, P., Sairanen, A., Nousiainen, J., Ja¨rvenranta, K., 2001. Laiduntamisen tahokkuuteen vaikuttavat tekija¨t (in Finnish with English absract). In: Niemela¨inen, O., Topi-Hulmi, M., Saarisalo, E. (Eds.), Nurmi 2001 symposium, esityksia¨ ja tilastokuvauksia. Suomen nurmiyhdistyksen julkaisu nro 15, pp. 42–48. Wardle, D.A., Yeates, G.W., Williamson, W.M., Bonner, K.I., Barker, G.M., 2004. Linking aboveground and belowground communities: the indirect influence of aphid species identity and diversity on a three trophic level soil food web. Oikos 107, 283–294. Wilsey, B.J., Coleman, J.S., McNaughton, S.J., 1997. Effects of elevated CO2 and defoliation on grasses: a comparative ecosystem approach. Ecological Applications 7, 844–853. Yeates, G.W., Bongers, M., de Goede, R.G.M., Freckman, D.W., Georgieva, S.S., 1993. Feeding habits in soil nematode families and genera—an outline for soil ecologists. Journal of Nematology 25, 315–331.