Skip to main content
NCBI home page
Annals of Botany logoLink to Annals of Botany
. 2008 May 7;102(1):113–125. doi: 10.1093/aob/mcn070

Genetic Identity Affects Performance of Species in Grasslands of Different Plant Diversity: An Experiment with Lolium perenne Cultivars

Christiane Roscher 1,2,*, Jens Schumacher 2,3, Wolfgang W Weisser 1, Ernst-Detlef Schulze 2
PMCID: PMC2712428  PMID: 18463110

Abstract

Background and Aims

Recent biodiversity research has focused on ecosystem processes, but less is known about responses of populations of individual plant species to changing community diversity and implications of genetic variation within species. To address these issues, effects of plant community diversity on the performance of different cultivars of Lolium perenne were analysed.

Methods

Populations of 15 genetic cultivars of Lolium perenne were established in experimental grasslands varying in richness of species (from 1 to 60) and functional groups (from 1 to 4). Population sizes, mean size of individual plants, biomass of individual shoots and seed production were measured in the first and second growing season after establishment.

Key Results

Population sizes of all cultivars decreased with increasing community species richness. Plant individuals formed fewer shoots with a lower shoot mass in more species-rich plant communities. A large proportion of variation in plant size and relative population growth was attributable to effects of community species and functional group richness, but the inclusion of cultivar identity explained additional 4–7 % of variation. Cultivar identity explained most variation (28–51 %) at the shoot level (biomass of individual tillers and reproductive shoots, seed production, heading stage). Coefficients of variation of the measured variables across plant communities were larger in cultivars with a lower average performance, indicating that this variation was predominantly due to passive growth reductions and not a consequence of larger adaptive plastic responses. No single cultivar performed best in all communities.

Conclusions

The decreasing performance of Lolium perenne in plant communities of increasing species richness suggests a regulation of competitive interactions by species diversity. Genetic variation within species provides a base for larger phenotypic variation and may affect competitive ability. However, heterogeneous biotic environments (= plant communities of different species composition) are important for the maintenance of intra-specific genetic variation.

Key words: Biodiversity, competition, genetic variation, growth reduction, Lolium perenne, phenotypic plasticity, species richness

INTRODUCTION

Over the last decades, the threat to biodiversity has raised concerns about the consequences of species loss for ecosystem functioning. A number of experimental studies have provided evidence that decreasing biodiversity can negatively affect ecosystem-level processes such as a decline in above-ground productivity or an increase in the number of invasive species (Hooper et al., 2005; Balvanera et al., 2006). Recent studies also explored the effects of plant community diversity on plant species themselves. In some of these studies, species were introduced into existing communities by seed addition or as transplanted individuals (Joshi et al., 2000; Symstad, 2000; Diemer and Schmid, 2001; Scherber et al., 2006; Mwangi et al., 2007). Other studies investigated the effects of community-level species richness on individual populations of plants within the community (Hector et al., 2002; van Ruijven and Berendse, 2003; Dimitrakopoulos and Schmid, 2004; Jumpponen et al., 2005; Roscher et al., 2007). Different responses of individual species, i.e. negative, neutral or positive effects of community species richness on species biomass production, indicate that species idiosyncrasy may be a key factor in these studies.

Plant performance does not only depend on environment, but also on genetic constitution. Genetic differentiation between plant populations in response to the abiotic or biotic environment is a common phenomenon (Bradshaw, 1984; Linhart and Grant, 1996). Genotype-specific responses to competition have been reported for several plant species (e.g. Turkington and Harper, 1979; Aarssen and Turkington, 1985; Kelley and Clay, 1987; Fridley et al., 2007). Natural selection that allows plant populations to evolve on ecologically relevant time scales (Thompson, 1998), and recent competition models among genotypes and species (Vellend, 2006) all emphasize the importance of genetic diversity for population performance, community composition and stability. Two main hypotheses propose causal, but opposite effects of community species richness on the range of genetic variation within a population (Vellend and Geber, 2005). The first hypothesis predicts a negative effect of community species richness on population-level genetic diversity because a diverse community of competing species restricts the chance for individual species to use different parts of a heterogeneous environment and only a subset of genotypes of a species may be able to coexist (stabilizing selection). The second hypothesis postulates a positive effect, because different local assemblages of species may favour different genotypes in competition (diversifying selection). So far, experimental studies on the relationship between population-level genetic diversity and community-level processes have been scarce. Booth and Grime (2003) demonstrated in a microcosm experiment with calcareous grassland species that community composition was more stable over a 5-year study period if component populations were made up of a larger number of genotypes.

Plant individuals are generally capable of phenotypic adaptations to the prevailing environmental conditions. Phenotypic plasticity is jointly determined by a genetic component, an environmental component and genotype–environment interactions (Schlichting, 1986; Pigliucci, 1996; Fordyce, 2006), but ultimately depends on the capacity of a genotype to produce different phenotypes under different environmental conditions (Via et al., 1995; Booy et al., 2000). Although the potential fitness benefits of phenotypic plasticity are generally accepted, trait plasticity does not necessarily have an adaptive value (Sultan, 1987). Phenotypic plasticity may evolve by genetic correlation with other traits under selection (Pigliucci et al., 2006). In addition, phenotypic responses may be the result of passive reductions in growth due to environmental stress such as resource limitation (Sultan, 1987, 2000; Dorn et al., 2000) that cannot be easily disentangled from active plastic responses (van Kleunen and Fischer, 2005; Valladares et al., 2007).

Despite several lines of evidence that genetic identity is important for species responses to plant community diversity, this issue has so far been largely ignored in biodiversity–ecosystem functioning research.

In the present study, experimental grasslands varying in species number (from 1 to 60) and number and identity of functional groups (from 1 to 4: grasses, small herbs, tall herbs, legumes) were used. Lolium perenne (Poaceae) was chosen as the model species and small populations of 15 different cultivars of this species were established in each experimental community. Lolium perenne is one of most important grass species in central Europe and worldwide used for the sowing and regeneration of temperate agricultural grassland. Its productivity is stimulated by nitrogen fertilization and it has a high fodder quality and grazing tolerance. Beside its importance as a fodder crop, perennial ryegrass is frequently used as a turf species in lawns (Beddows, 1967). There has been a large number of studies on physiological, morphological and phenological differences among L. perenne cultivars, including leaf length and tillering (Hazard et al., 1996; Gautier et al., 1999; Bahmani et al., 2000), growth rate and carbohydrate accumulation (Humphreys, 1989; Turner et al., 2001), heading and tiller turnover (Laidlaw, 2004), induction of flowering (Aamlid et al., 2000) and seed production (Elgersma, 1990). With the expectation of variation among L. perenne cultivars in the present model system, the main questions were as follows. (1) Do cultivars vary in their response to the richness of species and functional groups of local communities? (2) Do cultivars differ in the extent of phenotypic plasticity induced by different numbers and assemblages of plant communities? (3) Is the amount of phenotypic plasticity across plant communities, assessed as coefficient of variation, positively or negatively related to the overall performance of individual cultivars?

MATERIALS AND METHODS

Design of The Jena Experiment

The study was implemented as part of The Jena Experiment, a large integrated biodiversity study (Roscher et al., 2004). The experimental site is located in the floodplain of the river Saale in Jena (Thuringia, Germany, 50°55′N, 11°35′E, 130 m a.s.l.). Mean annual air temperature in the region is 9·3 °C, and average annual precipitation amounts to 587 mm (Kluge and Müller-Westermeier, 2000). The site was converted from grassland to an arable field around 1960 and received high fertilizer inputs for the last decades prior to the establishment of the experiment in 2002. The soil is a Eutric Fluvisol developed from up to 2 m thick fluvial sediments almost free of stones. Due to the fluvial dynamics of the Saale river, soil texture ranges from sandy loam near the river to silty clay with increasing distance from the river. Because of this gradient in soil characteristics, the site was divided into four blocks parallel to the river, such that effects of soil heterogeneity could be separated from biodiversity effects. Central European species-rich semi-natural grasslands traditionally mown twice a year (Arrhenatherion communities according to the vegetation classification of Ellenberg, 1988) served as the ‘target’ community to create a species pool for the experiment. Sixty species common to these grasslands were chosen and categorized into four functional groups: grasses (16 species), legumes (12 species), small herbs (12 species) and tall herbs (20 species) (Roscher et al., 2004). The experiment had a near-orthogonal design based on the factors species number (1, 2, 4, 8, 16, 60 species) and number of functional groups (1, 2, 3, 4 functional groups) with the limitation that it is not possible to have more functional groups than species in a mixture. Each species richness level was replicated 16 times except for the 16-species mixtures with 14 replicates because species number in two functional groups was too small to create mixtures containing legumes or small herbs only and the 60-species mixture with four replicates (Table 1). In total, 82 plots of 20 × 20 m size were established. Mixtures were created by drawing at random (with replacement). A complete species list of all experimental communities is given in Table S1 in the Supplementary Information, available online.

Table 1.

Summary of the experimental design showing the number of replicates for each combination of species number and functional group number

Species number
1 2 4 8 16 60
Functional group no.
1 16 8 4 4 2
2 8 4 4 4
3 4 4 4
4 4 4 4 4
No. of plots 16 16 16 16 14 4
Open in a new tab

Plots were sown from 11 to 16 May 2002. The numbers of seeds sown aimed at a density of 1000 seedlings m–2 divided equally among species and were corrected for germination rates following laboratory tests (Roscher et al., 2004). Plots were mown twice a year in early June and in the middle of September according to the typical management regime of extensively used hay-meadows. In the year of sowing, mowing first took place 8 weeks after sowing at the beginning of July. Plots were regularly weeded by hand to prevent establishment of unsown species. Plots remained unfertilized during the whole experiment. In the beginning of the experiment (autumn 2002) mineral NO3-N storage in the soil (30 cm depth) varied between 0·5 and 2·2 g m–2, while PO4-P storage was in a range 23·7–82·7 g m–2 (Oelmann, 2006). Although removal of mown plant material without fertilizer addition results in a continuous export of nutrients during the course of the experiment, nitrogen mineralization compensates N uptake by plants. Phosphorous supply is at an optimum level based on a classification used for CAL (calcium–lactate–acetate) extraction (Oelmann, 2006).

Sub-experiment with Lolium perenne cultivars

The grass Lolium perenne L. (perennial ryegrass) is a hemicryptophyte with a loose to densely tufted growth. Longevity of plant individuals ranges from annual or biennial life span to prevailing persistent forms. The individual tillers are annual or winter-annual, depending on the time taken to develop an inflorescence. Under suitable conditions plantlets may be formed in the leaf-axils and allow for clonal propagation. However, reproduction by seeds is the main mode of propagation (Beddows, 1967). Perennial ryegrass is a self-incompatible, wind-pollinated species. In common with many other temperate grasses, L. perenne requires low temperatures and/or short days for primary flower induction, and floral initiation depends on subsequent day length (Cooper, 1960; Heide, 1994).

Lolium perenne was not included in the pool of 60 species used to create the experimental communities because of the planned sub-experiment. Accessions of 15 cultivars of L. perenne were provided by the Institute of Plant Genetics and Crop Research Gatersleben (IPK, Germany). This selection covered a range of variability based on knowledge from agricultural cultivar standard tests on productivity, the capability to form dense swards and phenology (E. Willner, Institute for Plant Genetics and Crop Research Gatersleben, Germany, pers. comm.), but was restricted to diploid cultivars with a perennial life cycle. Accessions of seven forage cultivars (‘Barlano’, ‘Borvi’, Duramo', ‘Fennema’, ‘Gremie’, ‘Kerdion’, ‘Lilasso’) and accessions of eight turf cultivars (‘Bellatrix’, ‘Bianca’, ‘Entrar’, ‘Gator’, ‘Juwel’, ‘Langa’, ‘Lisuna’, ‘Lorina’) were included. Molecular analyses with SNP markers gave evidence that all cultivars used for this study were genetically different (S. Nestman, Max Planck Institute for Biogeochemistry, Jena, unpubl. res.). Preliminary germination tests served to control for differences in seed viability and to determine germination rates. Tests were performed on filter paper with three replicates per cultivar subjected to a standardized temperature and light regime (16 h day at 20 °C and a night temperature of 12 °C). Per replicate, 50 seeds were placed in a Petri dish and checked for germination every other day. Germination tests were terminated after 40 d.

The experiment with 15 Lolium cultivars was nested in the large-area plots of The Jena Experiment and was sown at the same time as the experimental communities. An area of 3 × 4 m in each 20 × 20 m plot was divided into 15 sub-plots of 0·8 × 1·0 m size. Cultivars were distributed randomly among these small sub-plots in each plot. The aimed sowing density amounted to 100 individuals m–2 (adjusted for average germination rates of 80 % from laboratory tests), in addition to the 1000 seedlings in the mixture of the plant community. In addition, in each experimental block a small-area monoculture of Lolium cultivars was established. Here, plots of 3·5 × 3·5 m were divided for the cultivars into sub-plots each 1/15th of the total plot area. The aimed sowing density was 1100 individuals m–2 to achieve a total plant density similar to the density in the experimental communities containing Lolium cultivars as additional species.

Measurements

The number of established Lolium individuals per sub-plot was counted after first mowing in July 2002 to determine initial population sizes 10 weeks after sowing. Counting of plant individuals was repeated 1 year later in July 2003. Relative growth rates of populations were calculated using the equation

graphic file with name mcn07007.jpg

where Nt1 is the number of individuals in the first year, Nt2 is the number of individuals in the second year, and a is the length of the time interval (= 1 year; Harper, 1977).

The number of vegetative tillers per Lolium individual was determined as a measure of plant individual size in early October in both years. Five individuals per cultivar and plot were selected at random, and the number of tillers per tussock was counted. The size of individuals with >100 tillers was estimated by counting the tillers in part of the tussock. Tillers of plantlets developing in leaf-axils which occurred in a few individuals (see above) were counted as tillers of the established plant individual. Counts per plot and cultivar were averaged. Mean relative growth rates of plant individuals were calculated according to the above equation using the numbers of tillers counted in both years.

The average size of flowering shoots and vegetative tillers was determined in the second year of the experiment in 2003. Flowering shoots were studied just before the first mowing between 4 and 10 June 2003 in mixtures with one, four and 16 species only. Eight randomly selected flowering shoots per cultivar and plot were harvested. They were classified on a five-part scale of flowering stages defined according to the relative position of the developing spike to the uppermost leaf of the flowering shoot: 0 = spike completely covered by the sheath of the uppermost leaf; 1 = spike conspicuously overtopped by the developing blade of the uppermost leaf, inflorescence not fully developed; 2 = top of spike and blade of the uppermost leaf of equal height; 3 = spike overtops blade of the uppermost leaf conspicuously, not flowering; 4 = flowering with visible anthers. The number of spikelets per spike was counted. The prevailing flowering stage of the flowering shoots was calculated as mode, whereas numbers of spikelets per spike were averaged per cultivar for each plot. All flowering shoots per sample were pooled to determine dry mass after drying at 70 °C for 48 h. Vegetative tillers were harvested between 14 and 19 August 2003 before the second mowing. Ten randomly selected tillers were cut off near the base, dead leaf material was separated, and samples were weighed after drying. The obligate vernalization for flower induction in L. perenne ensured that cultivars could be unequivocally assigned to their sub-plots during the first and the second growing season of the experiment. Populations consisted of the initially sown cohort because inflorescence development started almost exclusively after the first winter period.

Data analyses

Data were analysed using S-Plus® 7·0 (Insightful Corp., 2005). Because the experimental factors species number and functional group number are linked in The Jena Experiment (Roscher et al., 2004), it is not possible to completely disentangle their relative effects in a statistical analysis. Therefore, general linear models were used with sequential sum of squares (type I SS) following the a priori hypotheses of the experiment which addresses: sown species richness (partitioned into log-linear term and deviation from linear); functional group richness (partitioned into linear term and deviation from linear); and cultivar identity and interactions between cultivar and the experimental factors. Additionally, alternative statistical models were fitted that included separate contrasts for the presence/absence of each functional group and their interaction with Lolium cultivars to test for effects of particular functional groups. Block identity was incorporated as a factor to remove variance caused by a gradient of soil conditions in the experimental field (Roscher et al., 2004), but a block effect could also result from a specific order of plot management (weeding, mowing) or the block-wise organization of data collection. Because of the split-plot design of the Lolium cultivar experiment, the diversity terms (species richness, functional group richness, presence of particular functional groups) were tested at the between-plot level, while Lolium cultivars and their interaction with diversity terms were tested at the within-plot level. The following response variables were analysed: population size (number of individuals in the first and relative population growth in the second year, respectively), mean size of plant individuals (average number of tillers per tussock in the first and relative change in the second year, respectively), mean dry mass per vegetative tiller and per flowering shoot, flowering stage and number of spikelets per spike in the second year. In addition, the product of population size (number of plant individuals) and plant individual biomass (average tussock size multiplied by average dry mass per vegetative tiller) were used to estimate population performance (= estimated yield) from the second-year data. Analyses of variables characterizing generative reproduction included plots with one, four and 16 species (in total 46 plots), because these variables were not measured in all plots. By contrast, analyses of other variables were based on all species-richness levels (in total 82 plots). When necessary, the response variables were log-transformed to homogenize variances. Multiple comparisons (e.g. between specific cultivars) were not performed after the global split-plot analysis of variance because this study was mainly interested in the estimation of the relative contribution of the different explanatory factors, especially of the genetic variation between all the cultivars, to the different response variables. The hypotheses tested in the split-plot analysis of variance are well-defined a priori hypotheses and a Bonferroni-type correction was therefore not applied (Quinn and Keough, 2002).

The sequential sums of squares were used to estimate the amount of variation explained (= fraction of total sum of squares) by cultivar identity, the diversity treatments (sum of variance explained by community species richness and functional group richness) and differences among the cultivars in their response to the diversity treatments (interaction of cultivar identity × diversity treatments). For these calculations, a reduced data set of 46 plots (only mixtures with one, four and 16 species) was used for all variables to make the analyses comparable.

In addition, the performance (= estimated yields) of Lolium cultivars in mixtures (Pmix) was compared with the performance expected from the average of the four replicated monocultures of the respective cultivar (Pmono). Monoculture performance was corrected for sowing density (monocultures, 1100 Lolium individuals m–2; mixtures, 100 Lolium individuals m–2 and 1000 plant individuals of other species). A ratio (Pmix/Pmono) >1 indicates that performance in mixtures was better than performance in monocultures, while (Pmix/Pmono) <1 suggests the opposite.

Coefficients of variation (CVs) were computed across all experimental communities for estimated yield and its components to compare the extent of variation within each Lolium cultivar induced by different environments in species mixtures. Coefficients of variation across cultivars in each plant community were used to assess whether this variation changes in relation to community species richness.

RESULTS

Population size

Lolium cultivars differed significantly in the number of established individuals (Table 2). Initial population size in individual cultivars, measured 10 weeks after sowing, decreased with a log-linear increase of sown species richness (Fig. 1A) and was affected by the presence of particular functional groups. While the presence of tall herbs positively affected the establishment success of Lolium cultivars, the overall negative effects of other grasses in the experimental communities on initial population sizes depended on cultivar identity (significant interaction term; Table 2). Relative growth rates of populations estimated 1 year later decreased with a log-linear increase of community species richness and were often negative in mixtures with higher species richness (Table 2 and Fig. 1B). Thus, the negative effects of species richness on population sizes of Lolium cultivars increased from the first to the second year. However, relative population growth rates and the size of the species richness effect differed significantly among cultivars. Additionally, effects of the presence of legumes on changes in population size varied with Lolium cultivar identity.

Table 2.

Summary of statistical analyses of measures of vegetative growth in relation to the experimental treatments

No. of individuals
 Rel. population growth
 Tussock size
 Rel. tussock growth
 Dry mass per tiller
 Population performance
First year
 Second year
 First year
 Second year
 Second year
 Second year
Source of variation d.f. MS F MS F MS F MS F MS F MS F
Block 3 227·5 0·90 8·13 2·62 0·42 0·45 8·89 2·28 6457·4 6·42*** 78·77 1·99
SR (log-linear) 1 2981·9 11·82*** ↓ 40·71 13·11*** ↓ 9·04 9·80** ↓ 104·13 26·70*** ↓ 13247·5 13·18*** ↓ 1302·06 32·89*** ↓
Deviation from SR 4 303·8 1·20 4·26 1·37 0·95 1·03 5·38 1·38 694·5 0·69 69·18 1·75
FG (linear) 1 68·2 0·27 4·08 1·32 0·56 0·61 4·34 1·11 2·6 <0·01 12·86 0·32
Deviation from FG 2 257·8 1·02 7·71 2·48 0·66 0·71 3·80 0·97 815·2 0·81 19·72 0·61
Grasses 1 2030·9 8·97** ↓ 1·35 0·43 1·66 1·82 14·70 3·93 3055·6 3·13 26·50 0·67
Legumes 1 1·5 <0·01 6·47 2·12 5·47 6·39** ↓ 11·03 2·91 3114·1 3·20 22·05 0·55
Small herbs 1 2·3 <0·01 5·29 1·72 2·09 2·31 18·27 4·95* ↑ 412·3 0·41 20·32 0·51
Tall herbs 1 1945·3 8·54** ↑ 0·66 0·21 6·26 7·41** ↑ 14·01 3·73 316·9 0·31 29·35 0·74
Plot 70 252·2 3·10 0·92 3·90 1005·1 39·59
Cultivar 14 1462·3 74·54*** 2·36 12·63*** 0·34 15·22*** 1·93 6·73*** 6978·1 55·70*** 53·49 24·24***
Cultivar × SR (log-linear) 14 29·1 1·48 0·64 3·41*** 0·05 2·09** 1·30 4·52*** 70·1 0·56 12·79 5·80***
Cultivar × deviation from SR 56 21·2 1·08 0·28 1·47* 0·03 1·24 0·52 1·81*** 160·7 1·28 3·12 1·41*
Cultivar × FG (linear) 14 23·9 1·22 0·20 1·05 0·03 1·44 0·19 0·67 118·9 0·95 0·98 0·44
Cultivar × deviation from FG 28 22·2 1·13 0·18 0·96 0·02 1·09 0·27 0·94 83·7 0·67 1·38 0·63
Cultivar × grasses 14 39·2 2·03* 0·29 1·55 0·02 0·74 0·17 0·60 45·0 0·36 1·42 0·64
Cultivar × legumes 14 20·7 1·06 0·33 1·77* 0·02 0·79 0·32 1·12 101·0 0·80 2·40 1·09
Cultivar × small herbs 14 15·8 0·80 0·06 0·32 0·01 0·46 0·37 1·31 137·3 1·10 2·53 1·15
Cultivar × tall herbs 14 20·3 1·03 0·11 0·60 0·01 0·52 0·15 0·53 87·3 0·70 1·91 0·86
Residuals 1022 19·6 0·19 0·02 0·29 125·3 2·21
Open in a new tab

Due to the split-plot design of the experiment, species richness (SR) and functional group richness (FG) were tested at plot level, while cultivar and the interaction with the diversity treatments (SR, FG) were tested against the residuals. Note, that the terms for the presence of functional groups (grasses, legumes, small herbs, tall herbs) and their interaction with cultivar idenity were added alternatively in additional models. The analyses included mixtures of all species-richness levels (n = 82). Listed are the mean sums of squares (MS), F ratios (F) and the level of significance, where *P ≤ 0·05, **P < 0·01 and ***P < 0·001. Arrows indicate significant increase (↑) or decrease (↓) of the measures with community species richness or presence of particular functional groups.

Fig. 1.

Fig. 1.

Open in a new tab

Effect of plant species richness on (A) initial population size 2002 (= number of individuals 0·8 m–2), (B) relative population growth 2003, (C) initial tussock size 2002 (= average number of tillers per plant individual), (D) relative change of tussock size 2003, (E) dry mass per vegetative tiller 2003, and (F) estimated yield (population performance) 2003 of 15 cultivars of L. perenne. Points represent means per species-richness level.

Individual plant size

The mean number of tillers per tussock in the year of establishment (2002) and relative growth of plant individuals 1 year later (2003) differed significantly among Lolium cultivars. Tussock size and relative growth decreased with increasing community species richness, but the size of this effect varied among cultivars (Table 2 and Fig. 1C, D). On average, relative growth rates of Lolium tussocks were positive in one- to 16-species communities, whereas in communities made up of the full complement of 60 species Lolium tussocks had fewer tillers in the second year compared with the first year. In addition, it was observed that the number of tillers per tussock developed during the first growing season was positively affected by the presence of tall herbs, whereas the presence of legumes had negative effects on the average number of tillers per tussock. The presence of small herbs had significant positive effects on the relative growth of plant individuals in the second year (Table 2).

Average size of individual tillers (= tiller dry mass) in the second year differed significantly among Lolium cultivars. However, a log-linear increase of community species richness had negative effects on tiller dry mass in all cultivars (Table 2 and Fig. 1E).

Estimated yield per population

Estimated yield per population in the second year of the study (= number of plant individuals × average number of tillers × average tiller dry mass) depended on Lolium cultivar identity. Although all cultivars sustained a decrease of estimated yield with increasing species richness of the mixtures, the size of this effect varied between different Lolium cultivars (Table 2 and Fig. 1F).

Reproductive growth

Being exposed to low temperatures and short days during winter, nearly all surviving tillers of the first growing season induced inflorescence development in the spring of the second growing season. The mean flowering stage before the first mowing differed significantly among Lolium cultivars (Table 3 and Fig. 2A). The presence of legumes delayed inflorescence development of L. perenne independent of cultivar identity. Although dry mass of flowering shoots varied considerably among cultivars, overall negative effects of increasing community species richness were found on dry mass of flowering shoots (Table 3 and Fig. 2B). The average number of spikelets per spike differed significantly among cultivars (Fig. 2C). The negative effect of legume presence on the number of spikelets (Table 3) disappeared when the flowering stage was entered as covariate (analysis not shown). Thus, this effect was attributable to differences in the developmental stage of the spikes in mixtures with/without legumes. Community species richness had no effect on the flowering stage or the number of spikelets per spike.

Table 3.

Summary of statistical analyses of measures of reproductive growth in relation to the experimental treatments in the second year of the experiment

Flowering stage
 Dry mass per reproductive shoot
 No. spikelets
Source of variation d.f. MS F MS F MS F
Block 3 5·26 2·32 67033·0 2·96* 455·1 4·89**
SR (log-linear) 1 1·45 0·64 290213·5 12·81*** ↓ 137·2 1·47
Deviation from SR 1 0·86 0·38 9541·2 0·42 26·3 0·28
FG (linear) 1 0·23 0·10 53411·9 2·36 89·0 0·96
Deviation from FG 2 0·29 0·13 7741·9 0·34 2·6 0·03
Grasses 1 0·70 0·30 60562·7 2·80 0·8 <0·01
Legumes 1 20·77 11·83*** ↓ 31919·4 1·43 520·2 6·41* ↓
Small herbs 1 6·69 3·12 50016·7 2·29 112·0 1·21
Tall herbs 1 1·73 0·76 52495·2 2·41 212·9 2·37
Plot 37 2·27 22646·7 93·0
Cultivar 14 38·94 61·02*** 201483·9 38·64*** 986·2 39·63***
Cultivar × SR (log-linear) 14 0·88 1·38 7208·7 1·38 24·1 0·97
Cultivar × deviation from SR 14 0·42 0·66 3666·6 0·70 21·5 0·86
Cultivar × FG (linear) 14 0·61 0·96 4515·1 0·87 41·8 1·68
Cultivar × deviation from FG 28 0·89 1·40 8566·7 1·64* 43·9 1·76**
Cultivar × grasses 14 0·32 0·49 1925·0 0·36 12·4 0·49
Cultivar × legumes 14 0·89 1·41 7908·5 1·54 15·6 0·62
Cultivar × small herbs 14 0·61 0·96 4697·3 0·90 20·3 0·81
Cultivar × tall herbs 14 0·30 0·46 5231·3 1·00 19·9 0·79
Residuals 560 0·64 5214·9 24·9
Open in a new tab

Due to the split-plot design of the experiment, species richness (SR) and functional group richness (FG) were tested at plot level, while cultivar and the interaction with the diversity treatments (SR, FG) were tested against the residuals. Note, that the terms for the presence of functional groups (grasses, legumes, small herbs, tall herbs) and their interaction with cultivar identity were added alternatively in additional models. The analyses included mixtures with one, four and 16 species only (n = 46). Listed are the mean sums of squares (MS), F ratios (F) and the level of significance, where *P ≤ 0·05, **P < 0·01 and ***P < 0·001. Arrows indicate significant decrease (↓) of the measures with community species richness or presence of particular functional groups.

Fig. 2.

Fig. 2.

Open in a new tab

Effect of plant species richness on (A) flowering stage, (B) dry mass per flowering shoot, and (C) average number of spikelets per spike, measured before first mowing in early June 2003. Points represent means per species-richness level (mixture with one, four and 16 species only).

Relative population performance of cultivars in mixtures compared with cultivar monocultures (Pmix/Pmono)

The performance (= estimated yield) of Lolium cultivars in mixtures (Pmix) was compared with expected values derived from the average of the respective cultivar monocultures (Pmono). All cultivars reached an average Pmix/Pmono > 1 (inter-specific competition < intra-specific competition) in mixtures with one or two other species, although large variation within the species-richness levels was found. The ratio Pmix/Pmono decreased below or only slightly exceeded one (inter-specific competition > intra-specific competition) in mixtures with four and more other species. Averaged across all mixtures, the relative performance differed to some degree among Lolium cultivars (see Table S2 in Supplementary Information, available online).

Coefficients of variation (CVs) of the ratio Pmix/Pmono across cultivars were calculated for each mixture. These CVs increased significantly with species richness (F1,77 = 24·67, P < 0·001; Fig. 3), indicating a higher variation in competitive responses of Lolium cultivars with increasing inter-specific competition. Lolium cultivars were ranked for each mixture according to their relative population performance. The proportion of occurrences of these ranks is shown for all cultivars in Fig. 4 and illustrates that no cultivar performed best in all mixtures, but that each of them achieved its best performance in particular mixtures.

Fig. 3.

Fig. 3.

Open in a new tab

Effect of plant species richness on coefficients of variation (CVs) of relative population performance (Pmix/Pmono) across cultivars. CVs increased significantly with a log-linear increase of community species richness.

Fig. 4.

Fig. 4.

Open in a new tab

Ranking of L. perenne cultivars corresponding to their relative performance (Pmix/Pmono) in species mixtures (n = 82) in the second year of the experiment (2003). Ranks from 1 to 15 are given starting with the best cultivar in a mixture. Ranks between 2 and 5, 6 and 10 and 11 and 14 are summarized in the graph. The proportional occurrence of these ranks was calculated across all mixtures. Cultivars were arranged from left to right according to their overall relative success in species mixtures.

The magnitude of variation in population performance (= estimated yields) across all species mixtures, expressed as CV, was compared with the CVs of single components of population performance to inspect whether cultivars differed with respect to the traits responsible for the variation in estimated yields (Fig. 5A). CVs of population performance ranged from 0·99 to 1·59 for the 15 cultivars. In all cultivars, the largest variation across experimental mixtures was found for the size of the plant individuals (CV = 0·87 ± 0·14 s.d.), while coefficients of variation were lower for population size (CV = 0·50 ± 0·10 s.d.) and average dry mass per tiller (CV = 0·43 ± 0·08 s.d.). CV of population performance was linearly correlated with CVs of individual performance components. The significant negative correlation between CVs of population performance across mixtures and average relative population performance per cultivar (Pmix/Pmono) indicated that variation was particularly high in cultivars with a low average performance (Fig. 5B).

Fig. 5.

Fig. 5.

Open in a new tab

Coefficient of variation (CV) of population performance across all species mixtures in relation to (A) CVs of individual traits, and (B) in relation to relative population performance (Pmono/Pmix; mean ± 1 s.e.), for 15 L. perenne cultivars in the second year of the experiment (2003). Pearson correlation coefficients are given, where *P ≤ 0·05, **P < 0·01 and ***P < 0·001.

Relative importance of genetic variation

Although the diversity treatments (in particular species richness) and cultivar identity significantly affected the variables measured to characterize the performance of Lolium cultivars, the fraction of explained variation differed strongly (Fig. 6). Variation in characteristics of generative reproduction (flowering stage, number of spikelets per spike) was largely attributable to Lolium cultivar identity. Cultivar identity also explained a large percentage of variation in the biomass of vegetative tillers and reproductive shoots, but diversity treatments also accounted for a significant proportion of variation in these variables (see also Tables 2 and 3). By contrast, higher percentages of variation in population sizes and the size of individual plants (= tussock size) were explained by effects of the experimental treatments, although the establishment success (initial population sizes) varied considerably depending on cultivar identity. For all measured variables, differences among cultivars in response to the diversity treatments accounted for 3–6 % of variation.

Fig. 6.

Fig. 6.

Open in a new tab

Percentages of variance explained by diversity (sum of species richness and functional group richness), genetic variation (cultivar identity) and the interaction of diversity and genetic variation for several variables in populations of 15 L. perenne cultivars measured in monocultures and four- and 16-species mixtures (n = 46).

DISCUSSION

Effects of plant community diversity on cultivar performance

The results show that increasing species richness of the experimental communities had overall negative effects on the performance of 15 cultivars of L. perenne. These findings are in accordance with previous studies that reported that the positive biodiversity–productivity relationship at the community level is not necessarily related to a higher productivity of individual species in plant communities of increasing species richness (e.g. Naeem et al., 1996; Tilman et al., 1997; van Ruijven and Berendse, 2003; Roscher et al., 2007).

Plant species productivity is controlled via plant species density and plant individual size. Although density–size regulations have often been studied in populations of individual species (see Harper, 1977), the relationship between plant species density and plant individual size has attracted less attention in plant communities of higher species richness so far (but see Roscher et al., 2007). Studies with transplanted individuals in The Jena Experiment (Scherber et al., 2006; Mwangi et al., 2007) and at the Swiss BIODEPTH site based on a similar pool of grassland species (Diemer and Schmid, 2001) found no or negative effects of plant species richness on plant individual biomass. The present analysis of L. perenne sown with the same initial density in all experimental communities provided evidence for the occurrence of negative effects of plant species richness on plant performance at the population level (number of established and surviving individuals) and at the level of individual plants (tussock size, dry mass of vegetative tillers and flowering shoots). However, the negative effects of plant species richness were most pronounced with respect to the number of tillers per individual (Fig. 5A). Density experiments with L. perenne demonstrated that plant individual size of this species with a highly organized modular growth is predominantly determined by variation in tiller number, rather than by tiller size (Kays and Harper, 1974). Lolium perenne has a high tolerance to grazing and cutting. However, it is well known that lower light availability is mainly responsible for reduced tillering and increased leaf death in dense monocultures of perennial ryegrass, while growth control by nutrient availability is more important in stands of lower density (Ong, 1978; Simon and Lemaire, 1987). Previous biodiversity experiments have shown that more species-rich plant communities may utilize resources more completely by intercepting more light and taking up more nutrients (e.g. Spehn et al., 2005). Closed vegetation canopies are characterized by a gradient in light quantity (photon flux density) as well as changes in light quality (ratio of red to far red light) in the lower canopy levels (Monsi and Saeki, 1953; Jones, 1992). Lower red/far-red ratios change the phytochrome status and lead to a reduction in tillering rate in L. perenne (Deregibus et al., 1983). Self-shading at the tiller base in dense monoculture swards induces a reduction of tillering accompanied by an increasing weight of individual tillers in perennial ryegrass (Kays and Harper, 1974). In the present study, the reduced formation of tillers with increasing community species richness and in plant communities with legumes was probably due to shading by taller, more competitive species and could not be compensated for by an increased size in individual tillers. Smaller plant individuals have a lower chance of survival. Therefore, the more pronounced negative effects of plant species richness and legume presence on population sizes of L. perenne cultivars in the second year were not surprising. In particular, at the highest species-richness levels, a high mortality of Lolium individuals was found, suggesting a regulation of species diversity by competitive interactions.

The present study of L. perenne populations only included the first two years after the establishment of the experiment. A decrease in plant individual size with increasing plant species richness could in the longer term result in the extinction of L. perenne in these communities, as it is associated with a reduced seed production, and establishment from seed is the predominant mechanism of regeneration in this grass species. The neglible effects of the diversity treatments on the number of spikelets per spike imply that the number of seeds produced is mainly controlled by the number of flowering shoots per individual plant. Possible direct effects on seed quality could not be tested because differences in seed maturation prevented the development of ripe seeds in all cultivars before mowing. Generally, modules (= tillers) of L. perenne complete their life cycle with the development of an inflorescence, and smaller individuals thus inevitably produce a smaller number of flowering shoots. In addition, it was observed that the presence of legumes affected reproduction by a delay in inflorescence development. Spike emergence in L. perenne is influenced by the prevailing temperature conditions during floral development and stem elongation (Camlin, 1975). Thus, the effect of legumes on flowering stages could also be due to modified microclimatic conditions in dense plant canopies.

Phenotypic plasticity

Although the present study showed that the general outcome of competition across a species-richness gradient in different multi-species assemblages was similar for all cultivars tested, a significant proportion of variation in the variables measured was accounted for by cultivar identity (Fig. 6). In addition, the strength of the response to the species-richness gradient and the presence of particular functional groups in the experimental plant communities differed to some degree among cultivars (Table 2). It is well known that the degree of phenotypic plasticity in response to environmental conditions varies amongst individuals of a species and between traits (Schmid, 1992; Sultan, 1995; West-Eberhard, 2003). In plants, plasticity is often lower in association with reproductive structures than in association with vegetative growth characteristics (Schmid, 1992). In the present study, variables measured at the shoot level appeared to underlie stronger developmental constraints and had the lowest variation. In particular, variation in characteristics of generative reproduction (flowering stage, number of spikelets per spike) and biomass of individual shoots could largely be attributed to differences among cultivars and depended less on the resident plant community. Active phenotypic plasticity is a well-known mechanism of resource-use optimization to enhance plant performance. Therefore, a negative correlation between the degree of variation in population performance across the experimental mixtures and the average population performance of individual cultivars (Fig. 5B) suggests that this variation was not due to a plastic adaptation but caused by a passive reduction (‘non-adaptive’ plasticity) because of growth limitation (Scheiner, 1993; van Kleunen and Fischer, 2005; Ghalambor et al., 2007).

The importance of genetic variation

A number of studies have shown that competitive interactions among plant species favour particular genotypes and lead to a decline in genetic variation within plant populations, while a heterogeneous biotic environment has positive effects on within-species diversity (e.g. Turkington and Harper, 1979; Aarssen and Turkington, 1985; Kelley and Clay, 1987; Taylor and Aarssen, 1990; Lüscher et al., 1992; Vavrek, 1998). Only recently, Fridley et al. (2007) demonstrated that plant genotype performance is influenced by the genetic identity of neighbouring species, but that relationships among genotypes of different species may change with environmental conditions. The insurance hypothesis of Yachi and Loreau (1999) stresses the idea that species diversity provides greater guarantee of ecosystem functioning because species respond differently to environmental fluctuations. While previous studies on genotype-specific responses were mostly conducted in species-poor model systems, the present study with different cultivars of L. perenne across communities of different species richness and composition is consistent with the view that genetically based differences within species determine competitive ability in specific environmental situations or in communities with different competitors and that such differences may thus contribute to ecosystem stability (Aarssen 1989, 1992; Hughes and Stachowicz, 2004; Vellend, 2006; Whitlock et al., 2007). However, the increasing variation of relative population performance across cultivars with increasing plant community diversity (Fig. 3) supports the hypothesis that only a subset of genotypes of a species is able to coexist with a diverse community of competing species (Vellend and Geber, 2005).

In conclusion, differences in plant genetic constitution provide a basis for larger phenotypic variation and may increase competitive ability even in a subordinate species. However, the genotype-specific reponses to selection pressure caused by plant communities of different species composition imply that diversity in biotic environments is important for the maintenance of genetic diversity of a species.

SUPPLEMENTARY INFORMATION

Supplementary Information is available online at http://aob.oxfordjournals.org/ and consists of details of the 82 species mixtures, including numbers of species sown and functional groups (Table S1), and relative performance of Lolium cultivars calculated as the ratio of the performance in mixture (Pmix) to the performance expected from the average of the four replicated monocultures of the respective cultivar (Pmono; Table S2).

ACKNOWLEDGEMENTS

This study was supported by the Deutsche Forschungs-gemeinschaft (DFG-grant FOR 456). We thank the Institute for Plant Genetics and Crop Research Gater-sleben (Germany), in particular K. Bachmann and E. Willner, for providing seed material and information about Lolium cultivars. Many people are involved in The Jena Experiment; in particular, we gratefully acknowledge U. Wehmeier, H. Scheffler, S. Junghans, S. Eismann and L. Wacker for their patient assistance in preparation of seed material, sowing the seed for the Lolium experiment, and harvesting, counting and weighing the plant material. We thank M. Wagner (Sheffield) for constructive comments on an earlier version and for correcting the English. The comments of two anonymous referees helped to improve the manuscript.

LITERATURE CITED

  1. Aamlid TS, Heide OM, Boelt B. Primary and secondary induction requirements for flowering of contrasting European varieties of Lolium perenne. Annals of Botany. 2000;86:1087–1095. [Google Scholar]
  2. Aarssen LW. Competitive ability and species coexistence: a ‘plant's-eye’ view. Oikos. 1989;56:386–401. [Google Scholar]
  3. Aarssen LW. Causes and consequences of variation in competitive ability in plant communities. Journal of Vegetation Science. 1992;3:165–174. [Google Scholar]
  4. Aarssen LW, Turkington R. Within-species diversity in natural populations of Holcus lanatus, Lolium perenne and Trifolium repens from four different-aged pastures. Journal of Ecology. 1985;73:869–886. [Google Scholar]
  5. Bahmani I, Hazard L, Varlet-Grancher C, Betin M, Lemaire G, Matthew C, et al. Differences in tillering of long- and short-leaved perennial ryegrass genetic lines under full light and shade treatments. Crop Science. 2000;40:1095–1102. [Google Scholar]
  6. Balvanera P, Pfisterer AB, Buchmann N, He J-S, Nakashizuka T, Raffaelli D, Schmid B. Quantifying the evidence for biodiversity effects on ecosystem functioning and services. Ecology Letters. 2006;9:1146–1156. doi: 10.1111/j.1461-0248.2006.00963.x. [DOI] [PubMed] [Google Scholar]
  7. Beddows AR. Biological flora of the British Isles. Lolium perenne L. Journal of Ecology. 1967;55:567–587. [Google Scholar]
  8. Booth RE, Grime JP. Effects of genetic impoverishment on plant community diversity. Journal of Ecology. 2003;91:721–730. [Google Scholar]
  9. Booy G, Hendriks RJJ, Smulders MJM, van Groenendael JM, Vosman B. Genetic diversity and survival of populations. Plant Biology. 2000;2:379–395. [Google Scholar]
  10. Bradshaw AD. Ecological significance of genetic variation between populations. In: Dirzo R, Sarukhán J, editors. Perspectives on plant population ecology. Sunderland, MA: Sinauer Associates; 1984. pp. 213–228. [Google Scholar]
  11. Camlin MS. Variation in heading dates of perennial ryegrass cultivars as influenced by spring and early summer temperatures. Journal of the British Grassland Society. 1975;30:289–294. [Google Scholar]
  12. Cooper JP. Short day and low temperature induction in Lolium. Annals of Botany. 1960;24:232–246. [Google Scholar]
  13. Deregibus VA, Sanchez RA, Casal JJ. Effects of light quality on tiller production in Lolium spp. Plant Physiology. 1983;71:900–902. doi: 10.1104/pp.72.3.900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Diemer M, Schmid B. Effects of biodiversity loss and disturbance on the survival and performance of two Ranunculus species with differing clonal architectures. Ecography. 2001;24:59–67. [Google Scholar]
  15. Dimitrakopoulos PG, Schmid B. Biodiversity effects increase linearly with biotope space. Ecology Letters. 2004;7:574–583. [Google Scholar]
  16. Dorn LA, Pyle EH, Schmitt J. Plasticity to light cues and resource in Arabidopsis thaliana: testing for adaptive values and costs. Evolution. 2000;54:1982–1994. doi: 10.1111/j.0014-3820.2000.tb01242.x. [DOI] [PubMed] [Google Scholar]
  17. Elgersma A. Genetic variation for seed yield in perennial ryegrass (Lolium perenne L.) Plant Breeding. 1990;105:117–125. [Google Scholar]
  18. Ellenberg H. Vegetation ecology of central Europe. Cambridge: Cambridge University Press; 1988. [Google Scholar]
  19. Fordyce JA. The evolutionary consequences of ecological interactions mediated through phenotypic plasticity. Journal of Experimental Biology. 2006;209:2377–2383. doi: 10.1242/jeb.02271. [DOI] [PubMed] [Google Scholar]
  20. Fridley JD, Grime JP, Bilton M. Genetic identity of interspecific neighbours mediates plant responses to competition and environmental variation in a species-rich grassland. Journal of Ecology. 2007;95:908–915. [Google Scholar]
  21. Gautier H, Varlet-Grancher C, Hazard L. Tillering responses to light environment and to defoliation in populations of perennial ryegrass (Lolium perenne L.) selected for contrasting leaf length. Annals of Botany. 1999;83:423–429. [Google Scholar]
  22. Ghalambor CK, McKay JK, Carroll SP, Reznick DN. Adaptive versus non-adaptive phenotypic plasticity and the potential for contemporary adaptation in new environments. Functional Ecology. 2007;21:394–407. [Google Scholar]
  23. Harper JL. Population biology of plants. London: Academic Press; 1977. [Google Scholar]
  24. Hazard L, Ghesquière M, Barraux C. Genetic variability for leaf development in perennial ryegrass populations. Canadian Journal of Plant Science. 1996;76:113–118. [Google Scholar]
  25. Hector A, Bazeley-White E, Loreau M, Otway S, Schmid B. Overyielding in grassland communities: testing the sampling effect hypothesis with replicated biodiversity experiments. Ecology Letters. 2002;5:502–511. [Google Scholar]
  26. Heide OM. Control of flowering and reproduction in temperate grasses. New Phytologist. 1994;128:347–362. doi: 10.1111/j.1469-8137.1994.tb04019.x. [DOI] [PubMed] [Google Scholar]
  27. Hooper DU, Chapin FS, III, Ewel JJ, Hector A, Inchausti P, Lavorel S, et al. Effects of biodiversity on ecosystem functioning: a consensus of current knowledge. Ecological Monographs. 2005;75:3–35. [Google Scholar]
  28. Hughes AR, Stachowicz JJ. Genetic diversity enhances the resistance of a seagrass ecosystem to disturbance. Proceedings of the National Academy of Science of the USA. 2004;101:8998–9002. doi: 10.1073/pnas.0402642101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Humphreys MO. Water-soluble carbohydrates in perennial ryegrass breeding. I. Genetic differences among cultivars and hybrid progeny grown as spaced plants. Grass and Forage Science. 1989;44:231–236. [Google Scholar]
  30. Jones HG. Plants and microclimate: a quantitative approach to environmental plant physiology. Cambridge: Cambridge University Press; 1992. [Google Scholar]
  31. Joshi J, Matthies D, Schmid B. Root hemiparasites and plant diversity in experimental grassland communities. Journal of Ecology. 2000;88:634–644. [Google Scholar]
  32. Jumpponen A, Mulder CPH, Huss-Danell K, Högberg P. Winners and losers in herbaceous plant communities: insights from foliar carbon isotope composition in monocultures and mixtures. Journal of Ecology. 2005;93:1136–1147. [Google Scholar]
  33. Kays S, Harper JL. The regulation of plant and tiller density in a grass sward. Journal of Ecology. 1974;62:97–105. [Google Scholar]
  34. Kelley SE, Clay K. Interspecific competitive interactions and the maintenance of genotypic variation within two perennial grasses. Evolution. 1987;41:92–103. doi: 10.1111/j.1558-5646.1987.tb05773.x. [DOI] [PubMed] [Google Scholar]
  35. van Kleunen M, Fischer M. Constraints on the evolution of adaptive phenotypic plasticity in plants. New Phytologist. 2005;166:49–60. doi: 10.1111/j.1469-8137.2004.01296.x. [DOI] [PubMed] [Google Scholar]
  36. Kluge G, Müller-Westermeier G. Das Klima ausgewählter Orte der Bundesrepublik Deutschland: Jena. Berichte des Deutschen Wetterdienstes. 2000;213 [Google Scholar]
  37. Laidlaw AS. Effect of heading date of perennial ryegrass cultivars on tillering and tiller development in spring and summer. Grass and Forage Science. 2004;59:240–249. [Google Scholar]
  38. Linhart YB, Grant MC. Evolutionary significance of local genetic differentiation in plants. Annual Review of Ecology and Systematics. 1996;27:237–277. [Google Scholar]
  39. Lüscher A, Conolly J, Jacquard P. Neighbour specificity between Lolium perenne and Trifolium repens from a natural pasture. Oecologia. 1992;91:404–409. doi: 10.1007/BF00317630. [DOI] [PubMed] [Google Scholar]
  40. Monsi M, Saeki T. über den Lichtfaktor in den Pflanzengesellschaften und seine Bedeutung für die Stoffproduktion. Japanese Journal of Botany. 1953;14:22–52. [Google Scholar]
  41. Mwangi PN, Schmitz M, Scherber C, Roscher C, Schumacher J, Scherer-Lorenzen M, et al. Niche pre-emption increases with species richness in plant communities. Journal of Ecology. 2007;95:65–78. [Google Scholar]
  42. Naeem S, Håkansson K, Lawton JH, Crawley MJ, Thompson LJ. Biodiversity and plant productivity in a model assemblage of plant species. Oikos. 1996;76:259–264. [Google Scholar]
  43. Oelmann Y. Plant diversity effects on N and P cycling in an experimental grassland. Germany: Berlin University of Technology; 2006. PhD Thesis. [Google Scholar]
  44. Ong CK. The physiology of tiller death in grasses. 1. The influence of tiller age, size and position. Journal of the British Grassland Society. 1978;33:197–203. [Google Scholar]
  45. Pigliucci M. How organisms respond to environmental changes: from phenotype to molecules (and vice versa) Trends in Ecology and Evolution. 1996;11:168–173. doi: 10.1016/0169-5347(96)10008-2. [DOI] [PubMed] [Google Scholar]
  46. Pigliucci M, Murren CJ, Schlichting CD. Phenotypic plasticity and evolution by genetic assimilation. Journal of Experimental Biology. 2006;209:2362–2367. doi: 10.1242/jeb.02070. [DOI] [PubMed] [Google Scholar]
  47. Quinn GP, Keough MJ. Experimental design and data analysis for biologists. Cambridge: Cambridge University Press; 2002. [Google Scholar]
  48. Roscher C, Schumacher J, Baade J, Wilcke W, Gleixner G, Weisser WW, et al. The role of biodiversity for element cycling and trophic interactions: an experimental approach in a grassland community. Basic and Applied Ecology. 2004;5:107–121. [Google Scholar]
  49. Roscher C, Schumacher J, Weisser WW, Schmid B, Schulze E-D. Detecting the role of individual species for overyielding in experimental grassland communities composed of potentially dominant species. Oecologia. 2007;154:535–549. doi: 10.1007/s00442-007-0846-4. [DOI] [PubMed] [Google Scholar]
  50. van Ruijven J, Berendse F. Positive effects of plant species diversity on productivity in the absence of legumes. Ecology Letters. 2003;6:170–175. [Google Scholar]
  51. Scheiner SM. Genetics and evolution of phenotypic plasticity. Annual Review of Ecology and Systematics. 1993;24:35–68. [Google Scholar]
  52. Scherber C, Milcu A, Partsch S, Scheu S, Weisser WW. The effects of plant diversity and insect herbivory on performance of individual plant species in experimental grassland. Journal of Ecology. 2006;94:922–931. [Google Scholar]
  53. Schlichting CD. The evolution of phenotypic plasticity in plants. Annual Review of Ecology and Systematics. 1986;17:667–693. [Google Scholar]
  54. Schmid B. Phenotypic variation in plants. Evolutionary Trends in Plants. 1992;6:45–60. [Google Scholar]
  55. Simon JC, Lemaire G. Tillering and leaf area index on grasses in the vegetative stage. Grass and Forage Science. 1987;42:373–380. [Google Scholar]
  56. Spehn EM, Hector A, Joshi J, Scherer-Lorenzen M, Schmid B, Bazeley-White E, et al. Ecosystem effects of biodiversity manipulations in European grasslands. Ecological Monographs. 2005;75:37–63. [Google Scholar]
  57. Sultan SE. Evolutionary implications of phenotypic plasticity in plants. Evolutionary Biology. 1987;21:127–178. [Google Scholar]
  58. Sultan SE. Phenotypic plasticity and plant adaptation. Acta Botanica Neerlandica. 1995;44:363–383. [Google Scholar]
  59. Sultan SE. Phenotypic plasticity for plant development, function and life history. Trends in Plant Science. 2000;5:537–542. doi: 10.1016/s1360-1385(00)01797-0. [DOI] [PubMed] [Google Scholar]
  60. Symstad AJ. A test of the effects of functional group richness and composition on grassland invasibility. Ecology. 2000;81:99–109. [Google Scholar]
  61. Taylor DR, Aarssen LW. Complex competitive relationships among genotypes of three perennial grasses: implications for species coexistence. The American Naturalist. 1990;136:305–327. [Google Scholar]
  62. Thompson JN. Rapid evolution as an ecological process. Trends in Ecology and Evolution. 1998;13:329–332. doi: 10.1016/s0169-5347(98)01378-0. [DOI] [PubMed] [Google Scholar]
  63. Tilman D, Knops J, Wedin D, Reich P, Ritchie M, Sieman E. The influence of functional diversity and composition on ecosystem processes. Science. 1997;277:1300–1302. [Google Scholar]
  64. Turkington R, Harper JL. The growth, distribution and neighbour relationships of Trifolium repens in a permanent pasture. IV. Fine-scale biotic differentiation. Journal of Ecology. 1979;67:245–254. [Google Scholar]
  65. Turner LB, Humphreys MO, Cairns AJ, Pollock CJ. Comparison of growth and carbohydrate accumulation in seedlings of two varieties of Lolium perenne. Journal of Plant Physiology. 2001;158:891–897. [Google Scholar]
  66. Valladares F, Gianoli E, Gómez JM. Ecological limits to plant phenotypic plasticity. New Phytologist. 2007;176:749–763. doi: 10.1111/j.1469-8137.2007.02275.x. [DOI] [PubMed] [Google Scholar]
  67. Vavrek MC. Within-population genetic diversity of Taraxacum officinale (Asteraceae): differential genotype response and effect on interspecific competition. American Journal of Botany. 1998;85:947–954. [PubMed] [Google Scholar]
  68. Vellend M. The consequences of genetic diversity in competitive communities. Ecology. 2006;87:304–311. doi: 10.1890/05-0173. [DOI] [PubMed] [Google Scholar]
  69. Vellend M, Geber MA. Connections between species diversity and genetic diversity. Ecology Letters. 2005;8:767–781. [Google Scholar]
  70. Via S, Gomulkiewicz R, de Jong G, Scheiner SM, Schlichting CD, van Tienderen PH. Adaptive phenotypic plasticity: consensus and controversy. Trends in Ecology and Evolution. 1995;10:212–217. doi: 10.1016/s0169-5347(00)89061-8. [DOI] [PubMed] [Google Scholar]
  71. West-Eberhard MJ. Developmental plasticity and evolution. Oxford: Oxford University Press; 2003. [Google Scholar]
  72. Whitlock R, Grime JP, Booth R, Burke T. The role of genotypic diversity in determining grassland community structure under constant environmental conditions. Journal of Ecology. 2007;95:895–907. [Google Scholar]
  73. Yachi S, Loreau M. Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis. Proceedings of the National Academy of Science of the USA. 1999;96:1463–1468. doi: 10.1073/pnas.96.4.1463. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Annals of Botany are provided here courtesy of Oxford University Press

RESOURCES