SEARCH

SEARCH BY CITATION

Keywords:

  • Bromus madritensis;
  • comparative approach;
  • competition;
  • Crepis foetida;
  • facilitation;
  • fitness components;
  • plant trait

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    We tested the hypothesis that the competitive ability of a plant changes during its lifetime, by evaluating variations in the relative influence of standing biomass and litter on plant performance.
  • 2
    Seedlings of two annual herbaceous species (‘targets’), of contrasted life-forms (Crepis foetida and Bromus madritensis) were transplanted into an old-field. The competitive ability of seedlings, vegetative and reproductive plants towards neighbours was recorded over one growing season.
  • 3
    Fourteen traits related to plant morphology, growth and reproduction were measured to assess the competitive ability of targets. Relationships among traits were characterized to identify a set of traits as a surrogate of target competitive ability.
  • 4
    The two target species responded similarly. Early growth was facilitated by litter and, to a greater degree, by vegetative biomass. Thereafter, the effect of neighbouring vegetation on target performance was negative, with a maximum depressing effect on reproduction (especially seed number and date of flowering) despite a weak facilitative effect of litter. Plant basal diameter, measured at peak growth period, and total number of inflorescences per plant, were found to predict growth and reproductive components of competitive ability, respectively.
  • 5
    Assessing the effects of competition for population success from the response of vegetative plants will underestimate its importance because competition appears to exert its maximal impact on seed production.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Plant–plant interactions may change in both magnitude and direction (competition or facilitation) during a plant's lifetime (e.g. Gurevitch 1986; De Steven 1991a,b). However, most competition experiments focus on adult plants and, moreover, on growth responses (73% vs. 6% documenting effects on fecundity; see Aarssen & Keogh 2002 for a review of experimental studies). Lack of data on reproduction is probably due to the implicit assumption of a direct relationship between growth and reproduction under competition. However, the few available experimental data suggest the occurrence of more complex relationships between the components of fitness, with large consequences for the population dynamics of species. For example, the competitive abilities of survival and growth of some old-field species were found to differ between seedlings and adult plants, with the survival of seedlings correlated to plant abundance (Howard & Goldberg 2001). While the effect of litter on germination and seedling phases of an annual species was not demographically significant, its effect on adults led to a change in the shape of the recruitment curve and to reduce the carrying capacity of the populations (Molofsky et al. 2000). Therefore, the response of distinct fitness components to competition needs to be directly compared.

The two components of vegetation, litter and standing biomass, have distinct effects on plant performance. Litter can facilitate the early growth of targets by buffering lethal frosts (Heady 1956; Watt 1974), conserving water during drought (Fowler 1986) or adding nutrients (Facelli & Pickett 1991a). However, it can also limit the final biomass of plants by acting as a physical barrier (Facelli & Pickett 1991a), intercepting light (Goldberg & Werner 1983; Fowler 1986; Hamrick & Lee 1987; Carson & Peterson 1990; Facelli & Pickett 1991b) or producing toxics (Sydes & Grime 1981; Dejong & Klinkhamer 1985; Peterson & Facelli 1992; Foster & Gross 1997). On the other hand, standing biomass generally has a negative effect on plant performance after seedling establishment, because of competition for resources (e.g. Foster & Gross 1997; Suding & Goldberg 1999). Therefore, the balance between the relative influence of litter and of standing biomass on plant performance should vary during the lifetime of plants.

This study aimed at characterizing the relative influence of standing and litter biomass on target performance of seedlings, vegetative and reproductive plants. We carried out a field experiment in a Mediterranean old-field, dominated by the perennial grass Brachypodium phoenicoides (L.) Roem & Schult., transplanting seedlings of two annual target species, Bromus madritensis L. and Crepis foetida L. Non-destructive measures of plant morphology were used to monitor the behaviour of targets during their lifetime and 14 vegetative and reproductive traits were assessed to determine which trait or suite of traits is most important in affecting competitive ability at each life stage (Aarssen & Keogh 2002).

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

study site

The study was conducted between February and July 2003, in southern France, in an old-field located 30 km north of Montpellier (43°51′, 3°56′ E) in a subhumid Mediterranean climate. This field used to be a vineyard, which, following the removal of the vines, was abandoned 44 years before our study. Vegetation cover was dominated by a tussock perennial grass, Brachypodium phoenicoides, which represents about 95% of the standing biomass of the site (Garnier et al. 2004) and also produces a lot of litter, with parts of the dead leaves staying on erect tillers before falling to the soil.

target species and experimental design

Two annual species that dominate early successional stages in the area were selected to reflect differences in life-form: Bromus madritensis is a tillering grass and Crepis foetida is a composite that has vegetative rosettes and erect, leafy stems when flowering.

Seedlings of targets were transplanted into a randomized block design. Two rows of five 2 × 2 m experimental blocks were established in a homogenous part of the site. Bromus madritensis seedlings were transplanted into one of the rows, while the other row was used for Crepis foetida. Four experimental treatments (V−L−, both litter and living biomass removed; V−L+, only vegetation removed; V+L−, only litter removed; V+L+, unmanipulated vegetation) were applied to 100 × 100 cm plots located at the four corners of each block. Neighbours were removed, as appropriate, before planting by applying a glyphosate herbicide solution (Roundup) in January, and collecting dead material after a week. Plots were periodically hand-weeded during the growing season. Litter, both on top of the soil and attached to erect tillers, was removed by hand in L− treatments. Living plant biomass in V+ treatments was transformed into a monoculture of Brachypodium phoenicoides by removing all other species by hand.

Four seedlings of each target species were transplanted per plot in early March 2003, separated from each other by 20 cm within treatment. Seeds were collected in the surroundings of the experimental site and germinated in an unheated glasshouse in January 2003. Approximately 10 days after germination, 500 seedlings were transferred into individual seedling plug containers containing a commercial potting soil. After a further 20 days, 80 seedlings of similar phenological stage to plants in the field were randomly selected for the experiment. The initial shoot biomass of each seedling was estimated from regressions of seedling biomass on maximum plant height (for Bromus madritensis) or on maximum basal diameter (for Crepis foetida), derived from a subsample (50 individuals per species) of the remaining seedlings.

measurements of environmental conditions in treatments

Photosynthetically active radiation (PAR) was measured above the vegetation and on the ground near targets with two replicates per plot, using an Accupar light interception device (Decagon Devices Inc., Pullman, Washington, USA). Measurements were made in May, corresponding to the peak period of production of leaves. Light availability was expressed as the percentage of PAR reaching the ground surface.

Soil humidity was measured in May by evaluating the water content (= 100 × dry soil mass/wet soil mass) of two soil samples per plot. Two series of 10-cm depth samples were collected 1 and 5 days, respectively, after a heavy rain. The fraction of water lost by the soil after 5 days (Hdy) for each plot was calculated after determining the water content of the two sets of samples.

Two 20 × 20 cm replicates of total vegetation were harvested per plot in May, separated into standing and litter biomass (VB and LB, respectively), oven-dried for 2 days at 60 °C and then weighed.

trait measurements

Fourteen traits were measured (Table 1), characterizing seedling and adult growth, reproduction and plant longevity. Early biomass of seedlings (EVS) was assessed 1 month after transplantation by measuring the maximum plant height for Bromus madritensis and maximum basal diameter for Crepis foetida. Plant basal diameter (D) was recorded in May, at the end of the period of peak production of leaves, as the maximum width of the plant. Specific leaf area (SLA, the ratio of leaf area and dry mass), leaf dry matter content (LDMC, the ratio of leaf dry and fresh mass) and leaf length (LL) were evaluated on leaves of each target plant in May. Water-saturated SLA and LDMC were determined on leaf blades, following the protocols described by Garnier et al. (2001). After measuring the length of leaf blades, their fresh mass was measured and their projected area determined with an area meter (Delta-T Devices, Cambridge, UK, model MK2) before oven-drying at 60 °C for at least 2 days and measuring dry mass. Half of the targets (two individuals per plot) were harvested when leaf traits were measured. Above-ground dry biomass (PPB, peak plant biomass) was measured after drying samples at 60 °C for at least 2 days.

Table 1.  Abbreviations, units and associated plant function of measured traits
TraitsAbbreviationUnitPlant function/Plant life stage
Early vegetative sizeEVScmEarly growth
Plant basal diameter at peak productionDcmGrowth
Peak production biomassPPBgGrowth
Leaf lengthLLcmGrowth
Specific leaf areaSLAm2 kg−1Growth
Leaf dry matter contentLDMCg mg−1Growth
Final production biomassFPBgGrowth
Reproductive fractionReFReproduction
Seed biomassSBgReproduction
Number of inflorescencesINbReproduction
Flowering dateFlDJulian daysReproduction, phenology
Plant life spanPLSdaysLongevity
Seed massSMgReproduction, dispersal
Germinative power of produced seedsGPGermination

The remaining target plants were used for evaluation of reproductive and phenological traits. The date when each plant started to flower was recorded (FlD, flowering date). In June, the number of inflorescences (INb, inflorescence number) was measured for each plant. Inflorescences were harvested from each plant when seeds were mature, but before dispersal, weighed and stored at ambient conditions in the laboratory in order to preserve seed germinative potential. Seeds produced by each individual were sorted and weighed (SB, seed biomass). Four subsamples of 25 matured seeds per individual were weighed to determine seed mass (SM). The germinative power (GP) of these subsamples was evaluated in Petri dishes at 20 °C for 15 days.

Daily visits to the field enabled us to record the date when each plant died (PLS, plant life span). When dead, each individual was harvested and weighed. Final plant biomass (FPB) was obtained by adding this value to the biomass of inflorescences. The reproductive fraction (ReF) was determined as the ratio of inflorescence biomass to final plant biomass.

data analysis

Two-way analyses of variance were performed for each target species to test the influence of block (random) and treatments (fixed) on target traits. Log-transformation of variables was performed when necessary. Final plant biomass was used as a covariable when testing the effect of treatments on reproductive traits. Post-hoc SNK tests were performed when appropriate.

To characterize the differences in environmental conditions among treatments, a principal components analysis (PCA) was performed on fraction of water lost by soil, percentage of transmitted PAR, and standing and litter biomass averaged per plot.

To detect relationships between traits, a PCA was carried out for each target species on mean values of traits per plot and treatment. In order to interpret coordinates of plots, fraction of water lost by soil, percentage of transmitted PAR, standing and litter biomass averaged per plot were used as supplementary variables in PCAs.

The competitive ability of trait i of species j measured in treatment t was estimated, calculating a ln Response Ratio (Hedges et al. 1999) as

  • image(eqn1 )

where Ti, j, t  and Ti, j,V−L are mean value of trait i for species j in treatment t and V−L−, respectively. Analyses of variance on lnRRs were carried out, taking into account the effects of treatments, species identity and of their interaction.

Statistical analyses were performed with the Statistica package (Version 6).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The four treatments corresponded to fairly different environmental conditions (Fig. 1). The first axis of PCA, accounting for 53% of variation, was defined positively by the percentage of light transmitted by vegetation, and negatively by the standing biomass. Average standing biomass was 206 and 0 g m−2 in V+ and V− treatments, respectively, and percentage of light transmitted by vegetation was 25 and 100%. The second axis, explaining 24% of variance, partially discriminated V−L+ from the other treatments because of lower loss of soil water (28 vs. 37%), and V+L+ from V+L− because of lower percentage of transmitted light (15 vs. 40%, respectively). Mean litter biomass was 287.5 g m−2 and 140 g m−2 in V−L+ and V+L+ treatments, respectively.

image

Figure 1. Principal component analysis for environmental conditions among treatments. (a) Contribution of environmental variables to the first two axes. Abbreviations: Hdy, percentage of humidity lost by soil after rain; LB, litter biomass per plot; VB, vegetation biomass per plot; %PAR, percentage of transmitted light at target level. (b) Components of plots with different treatments: V−L−, both vegetation and litter removed; V−L+, vegetation removed; V+L−, litter removed; V+L+, unmanipulated vegetation.

Download figure to PowerPoint

The two target species responded similarly to treatments (Tables 2 and 3). After 1 month of growth, seedlings were larger in V+L+ and smaller in V−L− but vegetative adults (especially B. madritensis) were larger in V− treatments than in other conditions. Leaves differed significantly among treatments for B. madritensis only: they were longer, and had higher SLA and lower LDMC, in the V+L+ treatment than in other plots. Seed production (number of inflorescences per plant, number of seeds per plant, seed biomass per plant) varied significantly among treatments according to the sequence V−L+ > V−L− > V+L+ > V+L−, even when the influence of plant biomass was taken into account (e.g. ancova for the number of inflorescences: F = 5.85, P < 0.05, and 28.04, P < 0.001, for B. madritensis and C. foetida, respectively). The reproductive fraction was significantly lowest in V+L− for B. madritensis and was stable for C. foetida among treatments. The onset of reproduction was delayed in V+L− compared with other treatments (significant only for C. foetida), but no variation was detected when the plant biomass was used as a covariable (ancova: treatment, F = 0.42, NS; PPB, F = 4.09, P < 0.05). Plant life span did not vary between treatments although the duration of reproduction (life span − onset of reproduction) decreased in the presence of living neighbours (data not shown). Finally, neither mass nor germinative potential of seeds varied among treatments.

Table 2.  Results of anovas for Bromus madritensis. Mean values of traits are given for each treatment. The hierarchy of response to treatments is given with SNK test results (alphabetical characters). (1)Kruskall-Wallis test instead of anova; NS = not significant. Abbreviations of traits as in Table 1
TraitStatisticV−L−V−L+V+L−V+L+
EVS12.38***10.5013.6714.3518.53
cbba
D11.12***22.229.912.913.8
aabb
PPB7.36**3.494.932.222.4
abacbc
LL3.35*11.2912.4111.3214.64
aabba
SLA23.8***2726.534.337.4
bbaa
LDMC17.6***220.2221.2186183.9
aabb
FPB7.29**8.6417.413.753.11
ababcc
ReF4.01*0.3220.2480.0870.186
aabab
SB20.4***2.544.020.250.38
aabb
INb24.90***264261
aabb
FlD2.76NS(1)163.4165.2169.2163.8
aaaa
PLS2.33NS140.6144.4137.3133.1
aaaa
SM0.87NS9E-041.1E-031.1E-031.4E-03
aaaa
GP1.56NS0.260.250.290.38
aaaa
Table 3.  Results of anovas for Crepis foetida. Mean values of traits are given for each treatment. The hierarchy of response to treatments is given with SNK test results (alphabetical characters); NS = not significant. Abbreviations of traits as in Table 1
TraitStatisticV−L−V−L+V+L−V+L+
EVS21.75***1616.21827.6
bbba
D8.18**27.329.617.926.6
aaba
PPB12.20***2.722.310.982.76
aaba
LL4.35*12.4212.4411.8316.61
bbba
SLA0.57NS15.9217.3818.4117.23
aaaa
LDMC1.10NS0.130.130.130.14
aaaa
FPB40.6***28.5041.263.8611.70
aacb
ReF0.96NS0.120.100.090.11
aaaa
SB32.4***3.294.060.351.24
aacb
INb53.3***1591982152
aacab
FlD5.09*163.6165.8169.6165
bbab
PLS0.21NS137.1137.5135.5139
aaaa
SM0.62NS3.2E-043.3E-043.1E-043.5E-04
aaaa
GP0.35NS0.570.580.590.62
aaaa

Bromus madritensis plants were separated according to size along the first axis of PCA. Plants grown without neighbours were heavy, ‘long-lived’ and had large basal diameter, and also produced a lot of inflorescences and seeds, while plants grown in competition were small plants and had high SLA leaves (Fig. 2). These effects correlate with differences in amount of transmitted light and standing biomass between V+ and V− treatments (Fig. 1). The second axis of PCA discriminated between plants in the two V+ treatments: V+L+ plants were typically large after 1 month of growth and/or had long leaves, whereas the V+L− plants were late-flowering. The peak production biomass was negatively correlated with the flowering date and SLA, and positively with plant diameter. No correlation was found between early vegetative biomass and adult biomass, or between traits related to plant size and leaf morphology.

image

Figure 2. Principal component analysis for Bromus madritensis traits among treatments. (a) Contribution of traits to the first two axes. • = active variable; □ = supplementary variable. Plant traits as in Table 1. Supplementary variables: LB, litter biomass per plot; VB, vegetation biomass per plot; Hdy, percentage of humidity lost by soil in days following rain; %PAR, percentage of transmitted light at target level. (b) Components of plots with different treatments: V−L−, both vegetation and litter removed; V−L+, vegetation removed; V+L−, litter removed; V+L+, unmanipulated vegetation.

Download figure to PowerPoint

The pattern of response to treatments of C. foetida (Fig. 3) was very similar to that of B. madritensis, although a lower percentage of variance was explained by the first two axes (31% and 19%, respectively, instead of 50% and 20%). The only differences between B. madritensis and C. foetida PCAs were due to variations in location of plant biomass and diameter at peak production periods on the circles of correlations.

image

Figure 3. Principal component analysis for Crepis foetida traits among treatments. (a) Contribution of traits to the first two axes. (b) Components of plots with different treatments. • = active variable; □ = supplementary variable. Abbreviations as in Fig. 2.

Download figure to PowerPoint

Response ratios of plant size varied through the growing season and among treatments for both target species (Fig. 4a,b). All treatments, but especially V+L+, were facilitative for seedlings, probably due to the large accumulation of biomass. The facilitative effect on seedlings was lower in V−L+ and depended on litter biomass per plot. For example, the litter biomass was significantly lower (158 g m−2 ± 39, n = 7) in plots where early growth of B. madritensis was facilitated than in plots where no facilitation occurred (lnRR not significantly different from 0) (344 g m−2 ± 70, n = 3). Plant diameter, peak production biomass and final production biomass were slightly increased by the V−L+ treatment, but were depleted in V+ treatments, especially V+ L−. Seed production (seed biomass and number of inflorescences) was more depressed by all treatments than peak and final biomass (Fig. 4c,d). By contrast, the treatments had no effect on seed mass for C. foetida and a slightly positive one for B. madritensis (Fig. 4c,d). Finally, plant life span was not changed by treatments whereas the flowering date of C. foetida targets was significantly delayed in the V+L− treatment (Fig. 4e,f). The two target species did not differ in response ratio (except seed mass: F = 6.51, P = 0.017), or in response to treatments (treatment–species interaction: 0.02 < F < 1.91).

image

Figure 4. Ln Response Ratio of traits among treatments for (a), (c), (e) Bromus madritensis and (b), (d), (f) Crepis foetida. (a) and (b), responses of ▴, early vegetative size (EVS); ▿, plant basal diameter at ‘peak’ production (D); •, ‘peak’ plant biomass (PPB); ○, final plant biomass (FPB); (c) and (d) responses of •, seed mass (SM); ○, seed biomass (SB); ▾, number of inflorescences (INb); (e) and (f) responses of •, flowering date (FlD); ○, plant life span (PLS). Abbreviations of treatments: V−L+, vegetation removed; V+L−, litter removed; V+L+, unmanipulated vegetation. Only half standard errors of means are shown for clarity. Dotted lines are for ln RR = 0. Note the different scales.

Download figure to PowerPoint

The effects of treatments on plant size are summarized in Fig. 5. The direction of the effect of treatments was fixed after the juvenile stage but the magnitude of the target response varied throughout the plant lifetime, with maximal effect on plant reproductive biomass.

image

Figure 5. A synthesis of treatment effects on target biomass measured or assessed from plant morphological traits. The diameter of circles is proportional to the mean biomass of Bromus madritensis targets at each stage. V−L−, both vegetation and litter removed; V−L+, vegetation removed; V+L−, litter removed; V+L+, unmanipulated vegetation. The biomass of each seedling and each juvenile plant was estimated from a regression of seedling biomass on the maximum plant height (H) obtained from a sample of 50 individuals: initial biomass of Bromus madritensis = 0.008H−0.038, R2 = 0.63**.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The two target species, despite differences in growth forms, indicated a similar shift in influence of neighbours as plants moved from seedling to adult life stages (Tables 1 and 2, Figs 2 and 3), as shown by the lack of significant effects of the identity of species, or their interaction with treatments, on response ratios (Fig. 4). From facilitative or neutral effects on early growth, neighbours became competitive for adult growth and reproduction, with a maximum influence on reproductive plants (Figs 4 and 5). This shift is due to the change in relative influence of litter and standing biomass on targets. The net positive effect of treatments on seedlings was ‘litter biomass dependent’ with a threshold of about 200 g m−2. Although the range of litter biomass was small in our experiment (118 g m−2 to 418 g m−2), this result is in agreement with data from other studies reviewed in Xiong & Nilsson (1999). Litter had an overall positive effect, whose importance increased over the life of targets. On the other hand, the effect of standing biomass switched from facilitative at the juvenile stage to competitive at adult stages, with increased influence on reproductive plants. The addition of litter modulated the negative effect of standing biomass on adult plants for C. foetida but not for B. madritensis, probably because of an erect growth form allowing plants to escape from the litter layer.

The balance between positive and negative effects of neighbours generally reflects trade-offs between abiotic factors (e.g. Bertness & Callaway 1994; Greenlee & Callaway 1996; Callaway & Walker 1997; Holmgren et al. 1997). For Holmgren et al. (1997), facilitation occurs under dry conditions when the improvement of plant water relations under the canopy exceeds the costs caused by lower light levels. Here, litter had a positive effect on water availability and no effect on light transmitted to targets at the adult stage (Fig. 1), probably explaining its overall positive effects on targets. Other indirect effects of litter, such as the rise of air humidity and the prevention of extreme temperature fluctuations, might have had some influence on seedling growth, occurring at the end of winter, and seed production, occurring during summer drought (Dejong & Klinkhamer 1988; Franco & Nobel 1989; Valientebanuet et al. 1991; Holmgren et al. 1997; Eckstein & Donath 2005). Light and water availabilities were depleted in the presence of standing biomass, especially with occurrence of litter (Fig. 1), explaining the competitive effect of live vegetation on targets (Fig. 4). The reason why reproductive plants responded more to treatments than vegetative plants remains unclear, but is probably due to competition for light and water becoming more intense and more asymmetric as plants grew up. However, differences between treatments were maintained when plant size effects were removed from analyses, suggesting the influence of other non-trophic environmental factor(s).

Although competitive ability is often defined in terms of biomass production, it is clear from this experiment that other plant traits, with large demographic consequences, also respond to occurrence of neighbours and must therefore be taken into account when searching for surrogates of competitive ability. Different sets of traits related to competitive ability of plants have been proposed on the basis of theoretical frameworks (Tilman 1990; Goldberg 1996). However, no unique combination has been documented so far. Plant basal diameter measured at vegetative peak production was strongly correlated to plant vegetative, final and reproductive biomass for both species (Figs 2 and 3) and can therefore be used as a surrogate to assess both the direction and amount of response of vegetative plants. This has already been proposed by Kazakou & Navas (2004) and Navas & Moreau-Richard (2005), but our results show that its predictive value can be extended to the direction of response of reproductive plants (Fig. 4a,b). On the other hand, the number of inflorescences, which is a non-destructive trait, can be used to detect both direction and amount of response of the reproductive component of competitive ability (seed biomass) (Fig. 4c,d). Finally, no surrogate of plant longevity could be proposed, as this trait did not vary under competition (Fig. 4e,f).

At the community level, the differences in direction and amount of competitive ability among sites with different relative importance of litter and standing biomass may act on plant species composition because of changes in population demography (e.g. Aarssen 1992; Foster 1999). Molofsky et al. (2000) showed, using a population dynamics model, that the effects of litter on adult biomass, but not those on seedlings, were demographically significant. In our study, however, litter modulated the competitive effect of neighbours on vegetative and reproductive biomass of targets: plant sizes and seed pools were larger without than with standing biomass, but larger with than without litter biomass (Fig. 5). The largest effect of competition on reproductive plants was associated with a significant, but lower, change in seed pools without affecting seed quality, estimated by seed mass and germinative potential (Tables 2 and 3). As regenerative processes are commonly thought to be more important in determining the distribution of species in space and time than the performance of established plants (Grubb 1977; Gross & Werner 1982), the fact that the seed production was strongly affected by the relative importance of litter and standing biomass must be taken into account in further studies. This is even more important as the direction of the competitive ability of a species can shift from facilitative at the recruitment phase (e.g. Fowler 1988; Greenlee & Callaway 1996; Suding & Goldberg 1999) to competitive in reproductive plants (our study).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The authors would like to thank Cécile Podeur and Elena Kazakou for field assistance, and José Escarré and Denis Vile for comments on a previous version of the manuscript. We acknowledge very valuable revision and comments by Timothy Howard, one anonymous reviewer and Lindsay Haddon. This work was partly funded by the EU ‘VISTA’ (Vulnerability of Ecosystem Services to Land Use Change in Traditional Agricultural Landscapes) programme (contract EVK2-2001-000356).

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aarssen, L. (1992) Causes and consequences of variation in competitive ability in plant communities. Journal of Vegetation Science, 3, 165174.
  • Aarssen, L. & Keogh, T. (2002) Conundrums of competitive ability in plants: what to measure? Oïkos, 96, 531542.
  • Bertness, M. & Callaway, R. (1994) Positive interactions in communities. Trends in Ecology and Evolution, 9, 191193.
  • Callaway, R. & Walker, L. (1997) Competition and facilitation: a synthetic approach to interactions in plant communities. Ecology, 78, 19581965.
  • Carson, W. & Peterson, C. (1990) The role of litter in an old-field community – impact of litter quantity in different seasons on plant species richness and abundance. Oecologia, 85, 813.
  • De Steven, D. (1991a) Experiments on mechanisms of tree establishment in old-field succession – seedling emergence. Ecology, 72, 10661075.
  • De Steven, D. (1991b) Experiments on mechanisms of tree establishment in old-field succession – seedling survival and growth. Ecology, 72, 10761088.
  • Dejong, T. & Klinkhamer, P. (1985) The negative effects of litter of parent plants of Cirsium vulgare on their offspring – autotoxicity or immobilization. Oecologia, 65, 153160.
  • Dejong, T. & Klinkhamer, P. (1988) Seedling establishment of the biennials Cirsium vulgare and Cynoglossum officinale in a sand dune area – the importance of water for differential survival and growth. Journal of Ecology, 76, 393402.
  • Eckstein, R.L. & Donath, T.W. (2005) Interactions between litter and water availability affect seedling emergence in four familial pairs of floodplain species. Journal of Ecology, 93, 807816.
  • Facelli, J. & Pickett, S. (1991a) Plant litter – its dynamics and effects on plant community structure. Botanical Review, 57, 132.
  • Facelli, J. & Pickett, S. (1991b) Plant litter – light interception and effects on an old-field plant community. Ecology, 72, 10241031.
  • Foster, B. (1999) Establishment, competition and the distribution of native grasses among Michigan old-fields. Journal of Ecology, 87, 476489.
  • Foster, B. & Gross, K. (1997) Partitioning the effects of plant biomass and litter on Andropogon gerardi in old-field vegetation. Ecology, 78, 20912104.
  • Fowler, N. (1986) Microsite requirements for germination and establishment of three grass species. American Midland Naturalist, 115, 131145.
  • Fowler, N. (1988) What is a safe site? Neighbour, litter, germination date, and patch effects. Ecology, 69, 947961.
  • Franco, A. & Nobel, P. (1989) Effect of nurse plants on the microhabitat and growth of cacti. Journal of Ecology, 77, 870886.
  • Garnier, E., Cortez, J., Billes, G., Navas, M., Roumet, C., Debussche, M. et al. (2004) Plant functional markers capture ecosystem properties during secondary succession. Ecology, 85, 26302637.
  • Garnier, E., Shipley, B., Roumet, C. & Laurent, G. (2001) A standardized protocol for the determination of specific leaf area and leaf dry matter content. Functional Ecology, 15, 688695.
  • Goldberg, D. (1996) Competitive ability: definitions, contingency and correlated traits. Philosophical Transactions of the Royal Society of London Series B -Biological Sciences, 351, 13771385.
  • Goldberg, D. & Werner, P. (1983) The effects of size of opening in vegetation and litter cover on seedling establishment of Goldenrods (Solidago Spp.). Oecologia, 60, 149155.
  • Greenlee, J. & Callaway, R. (1996) Abiotic stress and the relative importance of interference and facilitation in Montane Bunchgrass communities in Western Montana. American Naturalist, 148, 386396.
  • Gross, K. & Werner, P. (1982) Colonizing abilities of biennial plant species in relation to ground cover – implications for their distributions in a successional sere. Ecology, 63, 921931.
  • Grubb, P. (1977) Maintenance of species richness in plant communities – importance of regeneration niche. Biological Reviews of the Cambridge Philosophical Society, 52, 107145.
  • Gurevitch, J. (1986) Competition and the local distribution of the grass Stipa Neomexicana. Ecology, 67, 4657.
  • Hamrick, J. & Lee, J. (1987) Effect of soil surface topography and litter cover on the germination, survival, and growth of Musk Thistle (Carduus Nutans). American Journal of Botany, 74, 451457.
  • Heady, H. (1956) Changes in a California annual plant community induced by manipulation of natural mulch. Ecology, 37, 798812.
  • Hedges, L., Gurevitch, J. & Curtis, P. (1999) The meta-analysis of response ratios in experimental ecology. Ecology, 80, 11501156.
  • Holmgren, M., Scheffer, M. & Huston, M. (1997) The interplay of facilitation and competition in plant communities. Ecology, 78, 19661975.
  • Howard, T. & Goldberg, D. (2001) Competitive response hierarchies for germination, growth and survival and their influence on abundance. Ecology, 82, 979990.
  • Kazakou, E. & Navas, M.-L. (2004) Variation in intensity of competition along a Mediterranean successional gradient. Ecology, Conservation and Management of Mediterranean Climate Ecosystems, Proceedings of the X MEDECOS Conference, 26–30 April 2004 (eds M.Arianoutsou & V.P. Papanastasis), pp. 18 . Millpress, Rhodos, Greece.
  • Molofsky, J., Lanza, J. & Crone, E. (2000) Plant litter feedback and population dynamics in an annual plant, Cardamine Pensylvanica. Oecologia, 124, 522528.
  • Navas, M.-L. & Moreau-Richard, J. (2005) Can traits predict the competitive response of herbaceous Mediterranean species? Acta Oecologica, 27, 107114.
  • Peterson, C. & Facelli, J. (1992) Contrasting germination and seedling growth of Betula alleghaniensis and Rhus typhina subjected to various amounts and types of plant litter. American Journal of Botany, 79, 12091216.
  • Suding, K. & Goldberg, D. (1999) Variation in the effects of vegetation and litter on recruitment across productivity gradients. Journal of Ecology, 87, 436449.
  • Sydes, C. & Grime, J. (1981) Effects of tree leaf litter on herbaceous vegetation in deciduous woodland. 2. An experimental investigation. Journal of Ecology, 69, 249262.
  • Tilman, D. (1990) Constraints and tradeoffs – toward a predictive theory of competition and succession. Oïkos, 58, 315.
  • Valientebanuet, A., Vite, F. & Zavalahurtado, J. (1991) Interaction between the cactus Neobuxbaumia tetetzo and the nurse shrub Mimosa luisana. Journal of Vegetation Science, 2, 1114.
  • Watt, A. (1974) Senescence and rejuvenation in ungrazed chalk grassland in Breckland – significance of litter and of moles. Journal of Applied Ecology, 11, 11571171.
  • Xiong, S. & Nilsson, C. (1999) The effects of plant litter on vegetation: a meta-analysis. Journal of Ecology, 87, 984994.