Plant and Soil 259: 9–17, 2004. © 2004 Kluwer Academic Publishers. Printed in the Netherlands.
9
Nutrient addition affects AM fungal performance and expression of plant/fungal feedback in three serpentine grasses Danny J. Gustafson1,2,3 & Brenda B. Casper1 1 Department
of Biology, University of Pennsylvania, Philadelphia, PA 19104-6018, U.S.A. 2 Current Address: Department of Biology, The Citadel, Charleston, SC 29409, U.S.A. 3 Corresponding author∗
Received 25 November 2002. Accepted in revised form 10 September 2003
Key words: arbuscular mycorrhizal fungi, biotic soil community, negative feedback, serpentine grassland, species coexistence Abstract Plant/soil microbial community feedback can have important consequences for species composition of both the plant and soil microbial communities, however, changes in nutrient availability may alter plant reliance on mycorrhizal fungi. In this research, we tested whether plant/soil community feedback occurs and if increased soil fertility altered the plant/soil community interactions. In two greenhouse experiments we assessed plant and AM fungal performance in response to different soils (and their microbial communities), collected from under three co-occurring plants in serpentine grasslands, and nutrient treatments. The first experiment consisted of two plant species (Andropogon gerardii, Sorghastrum nutans), their soil communities, and three nutrient treatments (control, calcium, N-P-K), while the second experiment used three plant species (first two and Schizachyrium scoparium), their soil communities collected from a different site, and two nutrient treatments (control, N-PK). Plant/soil community feedback was observed with two of the three species and was significantly affected by nutrient enrichment. Negative Sorghastrum/soil feedback was removed with the addition of N-P-K fertilizer at both sites. Andropogon/soil feedback varied between sites and nutrient treatments, while no differential Schizachyrium growth relative to soil community was observed. Addition of N-P-K fertilizer to the nutrient poor serpentine soils increased plant biomass production and affected plant/soil community interactions. Calcium addition did not affect plant biomass, but was associated with significant increases in fungal colonization regardless of plant species or soil community. Our results indicate that nutrient enrichment affected plant/soil community feedback, which has the potential to affect plant and soil community structure.
Introduction Feedback between plants and soil microbes can have important consequences for species composition of both the plant and soil communities. Positive feedback occurs when the plant alters soil characteristics, including the soil microbial community, in a way that enhances the performance of that species compared to co-occurring species. Conversely, negative feedback occurs when the plant changes the soil biota such that the performance of that plant species decreases compared to others. Negative plant/soil community ∗ FAX No: (843) 953-7264.
E-mail:
[email protected]
feedback appears to be more widespread (Bever et al., 1997, 2002; Bruehl, 1987; Klironomos, 2002; Mills and Bever, 1998; van der Heijden et al., 1998a, b; van der Heijden, 2002) and has been implicated as an important mechanism driving early succession (Johnson et al., 1991; Van der Putten, 1993; Van der Putten and Peters, 1997). In general, negative feedback is believed to promote species diversity in plant and soil communities while positive feedback can either enhance (Molofsky et al., 2001) or reduce diversity (Bever, 1999; Bever et al., 2002). Microbes known to be involved in feedback include a variety of bacteria and fungi, some of which are clearly pathogenic and others whose relationship
10 to plants may vary along a mutualistic-parasitic continuum (Bever, 1994; Castelli and Casper, 2003; Johnson et al., 1997; Mills and Bever, 1998; Smith and Read, 1997; Westover and Bever, 2001; Westover et al., 1997). Arbuscular mycorrhizal (AM) fungi, whose roles are often mutualistic, may mediate feedback if the fungal species differ in the amount that they benefit different host plant species or show specificity to host plant species (Bever, 1999, 2002; Bever et al., 2002). Mycorrhizal mediated negative feedback has been observed in a tropical forested system (Kiers et al., 2000), a North Carolina old field community (Bever, 1994; Bever et al., 1997), and an eastern serpentine grassland (Nottingham), which is studied here (Castelli, 2001; Castelli and Casper, 2003). Because the beneficial effects of mycorrhizal fungi can be facultative, depending on soil nutrient levels (Anderson and Liberta, 1989, 1992; Anderson et al., 1994; Johnson, 1993; Johnson et al., 1997; Kitt et al., 1988; Schultz et al., 2001; Smith and Read, 1997), among other factors, the occurrence or importance of feedback might also depend on nutrient levels. In this study we examined two serpentine grasslands to determine (1) whether differential plant performance consistent with plant/soil feedback can be demonstrated when plants are grown in the greenhouse in soils collected from the field under different plant species, (2) whether nutrient limitations on plant growth are evident, and (3) whether the response of plants to different soils changes with the addition of nutrients. The use of soils, instead of fungal inocula specifically, incorporates other soil factors, including non-mycorrhizal microbes, and should represent the soil diversity that plants could experience in the field. Nutrient treatments included N-P-K fertilizer and, for one site, a calcium treatment as well. Serpentine soils are characterized by potentially toxic levels of Mg, Fe, Cr, and Ni as well as low levels of N, P and K, and it has been suggested that low Ca:Mg ratios may be a factor limiting plant growth (Brooks, 1987).
Materials and methods Study sites and species These experiments were conducted with plant material and soil collected from grasslands on the Goat Hill (experiment 1) and Nottingham (experiment 2) serpentine barrens located in Chester County, Pennsylvania, U.S.A. These sites encompass both
forest and some of the largest remaining tracts of eastern serpentine grasslands (Latham, 1993). Eastern serpentine grasslands historically occurred in scattered locations from Nova Scotia to one reported community in Georgia. Twenty-six serpentine grassland sites ≥2 ha that remain; approximately half of the total area of the eastern serpentine grasslands occur in southeastern Pennsylvania and northeastern Maryland (Gustafson et al., 2003). Maintenance of the eastern serpentine grasslands are a conservation concern because they are hotspots for rare and disjunct species and they have been shrinking in area due to forest encroachment, which is believed to be the result of fire suppression, and urbanization (Latham, 1993; Tyndall and Hull, 1999). The dominant species of these serpentine grasslands are perennial C4 grasses that are characteristic of the central Great Plains of North America (Great Plains Flora Association, 1986; Gustafson et al., 1999), of which we selected Andropogon gerardii Vitman, Schizachyrium scoparium (Michx.) Nash, and Sorghastrum nutans (L.) Nash for this study. Seeds of each plant species were randomly collected from >100 plants at the Nottingham serpentine grassland, thoroughly mixed by species and cold dry stratified at 4 ◦ C. The processing of each plant species for each experiment was as follows: chaff was removed, seeds were germinated in greenhouse flats containing sterilized sand, watered with sterilized diH2 O as needed, and grown under greenhouse condition. Two week old seedlings were randomly selected from these flats and used in both experiments. Plant species will be referred to by genus. To test whether nutrient addition affects the documented plant/AM fungal negative feedback, we used soils collected from the two different serpentine grasslands in the following experiments. In the first experiment, soil was collected at Goat Hill from under several Andropogon and Sorghastrum plants, over several hectares, by removing soil throughout the vertical rooting zone to parent material (ca. 15 cm deep) directly below each plant. These soils were pooled by the plant species under which they were collected, passed through a 1 cm sieve to remove rocks and plant material, mixed with sterilized sand/soil (1:4) to improve drainage and placed in 10 cm × 10 cm × 15 cm square pots. Three nutrient treatments consisted of 5 g of N(13%)-P(13%)-K(13%) slow release fertilizer, 3.5 g of calcium (gypsum - 20% calcium), or no nutrients mixed into each pot. The experimental design consisted of two plant species, two soil types, three nutrient treatments, and 12 replicates for a total of 144
11 pots. Two week old (4–5th leaf stage) seedlings were planted one per pot and maintained in the University of Pennsylvania Department of Biology greenhouses, watered daily, and supplemented with artificial light on cloudy days. Any seedlings dying within the first week were replaced. After 12 weeks, plants were harvested, divided into shoots and roots, and dried at 90 ◦ C for 96 h for measures of total dry biomass. Because AM had been implicated in negative feedback in previous research (Castelli, 2001; Castelli and Casper, 2003), we evaluated the performance of mycorrhizal fungi in the different treatments. Before drying, the roots from half of the replicates were stained and scored for percent arbuscular mycorrhizal colonization following McGonigle et al. (1990). Root length was determined using WinRhizo Pro (Regent Instruments Inc., Quebec, CA) and root length colonized was calculated by multiplying the percent colonization by the total root length. AM root length was used as an estimate of fungal performance in addition to percent colonization because it has been shown to correlate better with spore production (Douds and Schenck, 1990) and provides an indirect measure of fungal mass. The occurrence of plant/soil community feedback depends on soil biotic communities affecting their preferred hosts differently than other plants. Three-way ANOVAs (plant species, soil origin, nutrient treatment) using plant or AM fungal performance as dependent variables, showed significant main effects and some interactions. Therefore, two-way ANOVAs (soil origin, nutrient) were performed separately for Andropogon and Sorghastrum. Ryan-Einot-Gabriel-Welsch (REGWQ) multiple range tests were used to identify significantly different means following ANOVA. Plant biomass (log(X+1)) and root length (log(X)) were transformed to satisfy the assumptions of normality and equal variance prior to analyses. All statistical analyses were performed using SAS (SAS Institute Inc, NC, U.S.A.). We repeated the same experimental design for a second serpentine grassland (Nottingham) except that we deleted the calcium treatment which had no effect on plant growth in the previous experiment and added a third plant species (Schizachyrium scoparium), known to show feedback with AM fungi in previous studies (Castelli and Casper, 2003). Nutrient treatments consisted of a 5 g N-P-K addition and a no nutrient addition control. Experimental conditions, plant and AM fungal performance measurements, and
Figure 1. Andropogon gerardii total biomass and AMF community performance in Goat Hill soils. Significant differences in plant/fungal performance between plants in Andropogon (AG) and Sorghastrum (SN) soils are indicated by different letters (P ≤ 0.05).
data analyses followed those outlined in experiment one.
Results Experiment 1 – Goat Hill Serpentine Grassland Analyses of root, shoot, and total biomass produced similar patterns, therefore we will present only total biomass. As main effects, Andropogon plants were larger with the addition of N-P-K fertilizer (F2,73 =
12 Table 1. Summary two-way ANOVA tables testing soil community, fertilizer, and soil community by fertilizer interactions for each plant species separately. (A) Goat Hill and (B) Nottingham A. Species
Source
Total Biomass d.f F value
Andropogon
S F S∗ F error S F S∗ F error
1 2 2 73 1 2 2 65
Species
Source
Total Biomass d.f F value
Andropogon
S F S∗ F error S F S∗ F error S F S∗ F error
2 1 2 52 2 1 2 65 2 1 2 47
Sorghastrum
B.
Sorghastrum
Schizachyrium
P
Source
12.67 110.52 2.09
0.007 <0.0001 0.13
1.30 50.58 4.68
0.25 <0.0001 0.01
S F S∗ F error S F S∗ F error
P
Source
5.13 93.47 6.58
0.009 <0.0001 0.002
0.46 261.03 6.20
0.63 <0.0001 0.003
0.68 44.09 1.12
0.51 <0.0001 0.33
S F S∗ F error S F S∗ F error S F S∗ F error
110.52, P < 0.001; Table 1A), but not with calcium (Figure 1). Andropogon plants were larger when grown in Andropogon soil (F1,73 = 12.67, P < 0.001) in the calcium and control treatments, but this differential plant growth as a consequence of soil origin was removed with the addition of N-P-K fertilizer (Figure 1). AM fungal colonization, as a measure of fungal performance, was higher (F2,26 = 10.11, P < 0.001) in the calcium treatment relative to the N-P-K or nutrient control but not different between the two soils. Root length colonized was not different between Andropogon and Sorghastrum soils or nutrient treatments (Figure 1). Sorghastrum also exhibited greater total biomass (F2,65 = 50.58, P < 0.001; Table 1A) with N-PK fertilization compared to the calcium and control treatments (Figure 2). There was no significant soil effect, but a significant soil by nutrient interaction (F2,65 = 4.68, P < 0.05). This interaction indicated plants were smaller when grown in Sorghastrum soil with no added nutrients, but did not differ among soils
AMF Colonization d.f F value P 1 2 2 31 1 2 2 26
0.98 10.11 0.98
0.33 0.0006 0.38
0.46 19.40 0.87
0.50 <0.0001 0.43
AMF Colonization d.f F value P 2 1 2 29 2 1 2 30 2 1 2 26
2.28 10.68 4.03
0.12 0.002 0.02
3.47 1.65 2.30
0.04 0.20 0.11
5.35 0.10 0.24
0.01 0.74 0.78
with either calcium or N-P-K fertilizer (Figure 2). As with the Andropogon, percent AM fungal colonization increased (F2,26 = 19.40, P < 0.001) with the addition of calcium, but there was no effect of soils and no soil by nutrient interaction (Figure 2). Sorghastrum root length colonized showed no soil, nutrient, or soil by nutrient interaction. Experiment 2 – Nottingham Serpentine Grassland For the Nottingham site, Andropogon plants were larger when fertilized (F1,52 = 93.47, P < 0.001; Table 1B) and when grown in Sorghastrum and Schizachyrium soils (F2,52 = 5.13, P < 0.01). The soil by nutrient (F2,52 = 6.58, P < 0.01) interaction reflects the fact that for fertilized soils, plants were smaller in soil collected under Andropogon than in the other two soil types (Figure 3). Percent AM fungal colonization was generally lower (F1,29 = 10.68, P < 0.01) with N-P-K addition. Root colonization did not differ with soil origin, but the soil by nutrient (F2,29 = 4.03,
13
Figure 2. Sorghastrum nutans total biomass and AMF community performance in Goat Hill soils. Significant differences in plant/fungal performance between plants grown in Andropogon (AG) and Sorghastrum (SN) soils for each nutrient treatment are indicated by different letters (P ≤ 0.05).
P < 0.05) interaction reflected higher colonization rates in Andropogon soil compared to the other soils in the no fertilization treatment only (Figure 3). Root length colonized showed no significant soil or nutrient main effects but there was a significant soil by nutrient interaction (F2,29 = 5.29, P < 0.05). Without nutrient addition, root length colonized was higher in Andropogon soil compared to Sorghastrum and Schizachyrium soils, but with added nutrients the reverse was true (Figure 3).
Figure 3. Andropogon gerardii total biomass and AMF fungal community performance relative to nutrient addition and soil biotic communities field collected under Andropogon (AG), Sorghastrum (SN), and Schizachyrium (SS). Significant differences in plant/fungal performance among plants grown in different soil communities for each nutrient treatment are indicated by different letters (P ≤ 0.05).
Sorghastrum plants were larger (F1,65 = 261.02, P < 0.001; Table 1B) when grown with N-P-K fertilizer (Figure 4). There was no effect of soil origin, but the soil by nutrient (F2,65 = 6.20, P < 0.01) interaction indicates plants were slightly smaller in their own soil only in the absence of nutrient enrichment (Figure 4). In general, Sorghastrum percent AM fungal colonization was lowest (F2,30 = 3.47, P < 0.05) when grown in its own soil and was not affected by nutrient addition. There was no soil
14
Figure 4. Sorghastrum nutans total biomass and AMF fungal community performance relative to nutrient addition and soil biotic communities field collected under Andropogon (AG), Sorghastrum (SN), and Schizachyrium (SS). Significant differences between plant/fungal performance among plants grown in different soil communities for each nutrient treatment are indicated by different letters (P ≤ 0.05).
by nutrient interaction. The different soil communities in which Sorghastrum was grown did not affect root length colonized, while fertilization (F1,30 = 16.76, P < 0.001) increased root length colonized. The significant soil by nutrient interaction (F2,30 = 3.59, P < 0.05) indicated less root length colonized when grown in Sorghastrum soil with no nutrient enrichment (Figure 4).
Figure 5. Schizachyrium scoparium total biomass and AMF fungal community performance relative to nutrient addition and soil biotic communities field collected under Andropogon (AG), Sorghastrum (SN), and Schizachyrium (SS). Significant differences between plant/fungal performance among plants grown in different soil communities for each nutrient treatment are indicated by different letters (P ≤ 0.05).
The N-P-K fertilizer increased Schizachyrium total biomass (F1,47 = 44.09, P < 0.001; Table 1B) while neither soil nor soil by nutrient interaction was significant (Figure 5). Schizachyrium AM fungal colonization (F2,26 = 5.35, P < 0.05) was higher in Schizachyrium soil compared to Sorghastrum soil, while nutrient and soil by nutrient interaction did not affect colonization rates (Figure 5). Schizachyrium
15 root length colonized (F1,26 = 10.62, P < 0.01) was higher with N-P-K addition, but did not differ by soil and there was no soil by nutrient interaction (Figure 5).
Discussion Our results indicate plant/soil community feedback for two of the three plant species grown in field-collected soil with no nutrient enrichment. Negative feedback between Sorghastrum and the soil community was observed with soils from both sites, while negative feedback was observed with Andropogon at Nottingham and positive feedback at Goat Hill. These results may help explain the natural distribution of these grasses in the field. Sorghastrum tends to be distributed thoughout the grassland, mixed with other grasses, while Andropogon occurs most frequently in the ecotone between the grassland and forest. Plant/soil community feedback does not appear to be mediated solely by AM fungi. This is because fungal performance did not always differ between soil types in ways that would explain the observed differential plant growth. If the negative feedback between Sorghastrum and the soil community was due exclusively to AM fungi, we would predict higher fungal performance and lower plant biomass when grown in its own soil community but lower fungal performance and higher plant biomass in other soils. Likewise, positive feedback involving mycorrhizae should result in increased plant and fungal performance in its own soil community relative to performance with a different soil community. Nutrient (N-P-K) addition seldom affected AM fungal colonization, which was consistent with results of three studies with Andropogon and Schizachyrium in the tallgrass prairie (Anderson and Liberta, 1992; Hetrick et al., 1986; Kitt et al., 1988). These results contrast with other studies that have shown a decrease in AM fungal colonization with an increase in nutrient levels, especially phosphorus, presumably because the plant does not need to incur the carbon cost to the fungi to acquire sufficient mineral nutrients (reviewed in Smith and Read, 1997). The lack of a decrease in fungal colonization with nutrient addition in these grasses could be due to the host plant being unable to regulate fungal infection (Blee and Anderson, 2000), possibly due to aggressive fungi in the soil community (Graham and Abbott, 2000), or some combination of plant characters and fungal characters. Determining the specific mechanism for this lack of reduced fungal
community colonization with nutrient addition is beyond the scope of this study. However the fact that this phenomenon occurs with the same grass species in both serpentine and prairie systems suggests there might be a general ecological pattern to this response. The addition of N-P-K fertilizer did, however, affect relative plant performance for two of the three species in the different soil types and thus affected whether feedback occurred. Negative feedback in Sorghastrum was removed and negative feedback with Andropogon was only observed with the N-P-K enrichment. The decoupling of plant/soil feedback may be an important ecological consequence of anthropogenic nutrient enrichment. In other studies, anthropogenic nutrient enrichment has been shown to alter soil mycorrhizal spore communities (Johnson et al., 1992; Egerton-Warburton and Allen, 2002). Our study showed no evidence that calcium levels (and therefore calcium/magnesium ratios) limit the growth of these grasses on serpentine soils as has been suggested in the literature (Brooks, 1987), but does point to calcium limitation of fungal growth. In a series of experiments with Schizachyrium in the tallgrass prairie, Anderson and Liberta (1989, 1992) showed a similar pattern of increased AM fungal colonization associated with the addition of calcium. In these studies, the increased colonization of fungi resulted in increased calcium in plant tissues, which we did not measure here. In general, the calcium requirements by monocotyledon plants is lower than dicotyledon plants (Marschner, 1995), but the calcium levels required for optimal mycorrhizal fungal growth are apparently unknown. In a study of soil nutrient limitations on arbuscular mycorrhizal fungi growth, Treseder and Allen (2002) found that external hyphal colonization of plant roots by Scutellaspora spp. was higher with phosphorus than nitrogen fertilization. Thus, phosphorus may be a limiting resource in mycorrhizal fungal growth. Clearly, the widely held view that AM fungi are not limited by mineral nutrients (Smith and Read, 1997) needs to be reevaluated. Our results do not completely agree with previous studies of plant/soil feedback at the Nottingham site (Castelli, 2001). Prior studies conducted both in the field, where seedlings were planted in gaps created in the clumps of the different grasses, and in the greenhouse using root pieces as sources of fungal inocula showed negative feedback for Sorghastrum but not for Schizachyrium, just as in our study. The discrepancies come from Andropogon, which consistently showed negative feedback in those studies but
16 in our study only when nutrients were added. It is possible that nutrient levels in the prior studies were elevated by autoclaving the soil. However, negative feedback was also demonstrated with Andropogon in a field experiment with natural, non-sterilized soils. Differences in results between the field study and ours using non-sterilized field soil are more difficult to explain. However, much of the differential Andropogon growth among soils under different host plant species in the field can be attributed to higher mortality than we observed. In fact, under field conditions Andropogon mortality exceeded 80% when grown in the Andropogon soil community compared to 23% and 10% in Sorghastrum and Schizachyrium soil communities, respectively (Castelli, 2001). It is likely that a combination of environmental stress in nature (e.g., water and heat stress) and biotic stress imposed by its own soil community contributed to negative feedback in the field, while our Andropogon plants were grown under less stressful environmental conditions of the greenhouse and were able to overcome some of the negative impact of its own soil community. We also realize that our use of sterile sand to improve pot drainage in the pot study somewhat changed the soil composition. Given these different results we do not recommend that greenhouse studies alone be used as a proxy for studying the dynamics of plant/soil microbe community feedback in natural systems. The plant and soil microbe community performance can depend on host plant species, herbivory, non-pathogenic soil microbe community, mycorrhizal fungal interactions, and root pathogens (Bever et al., 1996; Gange et al., 2002; Hamilton and Frank, 2001; Hart and Reader, 2002; Klironomos, 2002; Mills and Bever, 1998; Pearson and Jakobsen, 1993; van der Heijden et al., 1998a,b; Westover and Bever, 2001; Westover et al., 1997), therefore we suggest that plant/soil microbe community feedback studies use some combination of field and greenhouse experiments with organisms that occur within the same plant community.
Acknowledgements We gratefully acknowledge the assistance of Dr Lena Cherkashina for assistance in the field and greenhouse, J. Klemens, L. Spindler, R. Lucas, K. Rowley, and two anonymous reviewers for providing comments on the manuscript. We would like to thank the Nottingham
County Park for allowing us access to the Nottingham field site.
References Allen M F 1996 The ecology of mycorrhizae. Cambridge University Press, Cambridge. Anderson R C and Liberta A E 1992 Influence of supplemental inorganic nutrients on growth, survivorship, and mycorrhizal relationships of Schizachyrium scoparium (Poaceae) grown in fumigated and unfumigated soil. Am. J. Bot. 79, 406–414. Anderson R C and Liberta A E 1989 Growth of little bluestem (Schizachyrium scoparium) (Poaceae) in fumigated and nonfumigated soils under various inorganic nutrient conditions. Am. J. Bot. 76, 95–104. Anderson R C, Hetrick B A D and Wilson G W T 1994 Mycorrhizal dependence of Andropogon gerardii and Schizachyrium scoparium in two prairie soils. Am. Midl. Natural. 132, 366–376. Bever J D 1994 Feedback between plants and their soil communities in an old field community. Ecology 75, 1965–1977. Bever J D 1999 Dynamics within mutualism and the maintenance of diversity: Inference from a model of interguild frequency dependence. Ecol. Lett. 2, 52–61. Bever J D 2002 Negative feedback within a mutualism: Hostspecific growth of mycorrhizal fungi reduces plant benefit. Proc. Royal Soc. London Ser. B 269, 2595–2601. Bever J D, Westover K M and Antonovics J 1997 Incorporating the soil community into plant population dynamics: the utility of the feedback approach. J. Ecol. 85, 561–573. Bever J D, Pringle A and Schultz P A 2002 Dynamics within the plant-arbuscular mycorrhizal fungal mutualism: testing the nature of community feedback. In Mycorrhizal Ecology. Eds. M G A van der Heijden and I R Sanders. pp. 267–292. Springer Verlag, New York. Bever J D, Morton J B, Antonovics J and Schultz P A 1996 Host-dependent sporulation and species diversity of arbuscular mycorrhizal fungi in a mown grassland. J. Ecol. 84, 71–82. Blee K A and Anderson A J 2000 Defense responses in plants to arbuscular mycorrhizal fungi. In Current Advances in Mycorrhizae Research. Eds. G K Podila and D D Douds Jr. pp. 27–44. APS Press, St. Paul. Brooks R T 1987 Serpentine and its vegetation: a multidisciplinary approach. Dioscorides Press, Portland. Bruehl G W 1987 Soilborne Plant Pathogens. Macmillan Publishing Co., New York. Castelli J P 2001 Negative mycorrhizal feedback and plant species coexistence in a serpentine grassland. Ph.D. thesis, University of Pennsylvania. Castelli J P and Casper B B 2003 Intraspecific AM fungal variation contributes to plant/fungal feedback in a serpentine grassland. Ecology 84, 323–336. Douds D D and Schenck N C 1990 Relationship of colonization and sporulation by VA mycorrhizal fungi to plant nutrient and carbohydrate content. New Phytol. 116, 621–627. Gange A C and Brown V K 2002 Action and interaction of soil invertebrates and arbuscular mycorrhizal fungi in affecting the structure of plant communities. In Mycorrhizal Ecology. Eds. M G A van der Heijden and I R Sanders. pp. 321–344. Springer Verlag, New York. Gange A C, Stagg P G and Ward L K 2002 Arbuscular mycorrhizal fungi affect phytophagous insect specialism. Ecol. Lett. 5, 11– 15.
17 Graham J H and Abbott L K 2000 Wheat responses to aggressive and non-aggressive arbuscular mycorrhizal fungi. Plant Soil 220, 207–218. Great Plains Flora Association 1986 Flora of the Great Plains, University of Kansas Press, Lawrence. Gustafson D J, Gibson D J and Nickrent D L 1999 Randomly amplified polymorphic DNA (RAPD) variation among remnant big bluestem (Andropogon gerardii Vitman) populations from Arkansas’ Grand Prairie. Mol. Ecol. 8, 1693–1701. Gustafson D J, Romano G, Latham R E and Morton J K (2003) Amplified fragment length polymorphism analysis of genetic relationships among the serpentine barrens endemic Cerastium velutinum Rafinesque var. villosissimum Pennell (Caryophyllaceae) and closely related Cerastium species. J. Torrey Bot. Soc. 130, 218–223. Hamilton E W and Frank D A 2001 Can plants stimulate soil microbes and their own nutrient supply: evidence from a grazing tolerant grass. Ecology 82, 2397–2402. Hart M M and Reader R J 2002 Taxonomic basis for variation in the colonization strategy of arbuscular mycorrhizal fungi. New Phytol. 153, 335–344. Hetrick B A D, Kitt D G and Wilson G T 1986 The influence of phosphorus fertilization, drought, fungal species, and nonsterile soil on mycorrhizal growth response in tallgrass prairie plants. Can. J. Bot. 64, 1199–1203. Jakobsen I, Smith S E and Smith F A 2002 Function and diversity of arbuscular mycorrhizae in carbon and mineral nutrition. In Mycorrhizal Ecology. Eds. M G A van der Heijden and I R Sanders. pp. 75–92. Springer Verlag, New York. Johnson N C 1993 Can fertilization of soil select less mutualistic mycorrhizae? Ecol. Appl. 3, 749–757. Johnson N C, Graham J H and Smith F A 1997 Functioning of mycorrhizal associations along the mutualism-parasitism continuum. New Phytol. 135, 575–585. Johnson N C, Zak D R, Tilman D and Pfleger F L 1991 Dynamics of vescular–arbuscular mycorrhizae during old field succession. Oecologia 86, 349–358. Johnson N C, Tilman D and Pfleger F L 1992 Plant and soil controls on mycorrhizal fungal communities. Ecology 73, 2034–2042. Kiers E T, Lovelock C E, Krueger E L and Herre E A 2000 Differential effects of tropical mycorrhizal fungal inocula on root colonization and tree seedling growth: implications for tropical forest diversity. Ecol. Lett. 3, 106–113. Kitt D G, Hetrick B A D and Wilson G W T 1988 Relationship of soil fertility to suppression of the growth response of mycorrhizal big bluestem in non-sterile soil. New Phytol. 109, 473–481. Klironomos J N 2002 Feedback with soil biota contributes to plant rarity and invasiveness in communities. Nature 417, 67–70. Latham R E 1993 The serpentine barrens of temperate eastern North America: critical issues in the management of rare species and communities. Bartonia 57(suppl.), 61–74. Marschner H 1995 Mineral nutrients of higher plants. Academic Press, New York. McGonigle T P, Miller M H and Evans D G 1990 A new method which gives an objective measure of colonization of roots by
vesicular-arbuscular mycorrhizal fungi. New Phytol. 115, 495– 501. Mills K E and Bever J D 1998 Maintenance of diversity within plant communities: soil pathogens as agents of negative feedback. Ecology 79, 1595–1601. Molofsky J, Bever J D and Antonvics J 2001 Coexistence under positive frequency dependence. Proc. Royal Soc. London Ser. B 268, 273–277. Pearson J N and Jakobsen I 1993 The relative contribution of hyphae and roots to phosphorus uptake by arbuscular mycorrhizal plants, measured by dual labelling with 32 P and 33 P. New Phytol. 124, 489–494. Schultz P A, Miller R M, Jastrow J D, Rivetta C V and Bever J D 2001 Evidence of a mycorrhizal mechanism for the adaptation of Andropogon gerardii (Poaceae) to high- and low-nutrient prairies. Am. J. Bot. 88, 1650–1656. Smith S E and Read D J 1997 Mycorrhizal symbiosis. Academic Press, San Diego. Treseder K K and Allen M F 2002 Direct nitrogen and phosphorus limitation of arbuscular mycorrhizal fungi: a model and field test. New Phytol. 155, 507–515. Tyndall R W and Hull J C 1999 Vegetation, flora, and plant physiological ecology of serpentine barrens of eastern North America. In Savannas, Barrens, and Rock Outcrop Plant Communities of North America. Eds. R C Anderson, J S Fralish and J M Baskin. pp. 67–82. Cambridge University Press, Cambridge. van der Heijden M G A 2002 Arbuscular mycorrhizal fungi as a determinant of plant diversity: in search for underlying mechanisms and general principles. In Mycorrhizal Ecology. Eds. M G A van der Heijden and I R Sanders. pp. 243–265. Springer Verlag, New York. van der Heijden M G A, Boller T, Wiemken A and Sanders I R 1998a Different arbuscular mycorrhizal fungal species are potential determinants of plant community structure. Ecology 79, 2082–2091. van der Heijden M G A, Klironomos J N, Ursic M, Moutoglis P, Streitwolf-Engel R, Boller T, Wiemken A and Sanders I R 1998b Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396, 69–72. Van der Putten W H and Peters B A M 1997 How soil-borne pathogens may affect plant competition. Ecology 78, 1785–1795. Van der Putten W H, Van Dijk C and Peters B A M 1993 Plantspecific soil-borne diseases contribute to succession in foredune vegetation. Nature 362, 53–56. Westover K M and Bever J D 2001 Mechanisms of plant species coexistence: roles of rhizosphere bacteria and root fungal pathogens. Ecology 82, 3285–3294. Westover K M, Kennedy A C and Kelley S E 1997 Patterns of rhizosphere microbial community structure associated with co-occurring plant species. J. Ecol. 85, 863–873. Section Editor: J.H. Graham
