SEARCH

SEARCH BY CITATION

Keywords:

  • complexity;
  • MacArthur;
  • microbial;
  • protists;
  • stability

Summary

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

1. Food-web complexity-stability relations are central to ecology, and many empirical studies show greater food-web complexity leads to lower population stability. Here, predator population variability decreased with increasing prey diversity in aquatic microcosm experiments, an example of greater food-web complexity leading to greater population stability.

2. Prey diversity as well as different sets of prey species within each level of prey diversity produced differences in predator population dynamics, demonstrating the importance of both prey composition and prey diversity in determining predator population stability.

3. Prey diversity can affect predator population dynamics through at least three groups of mechanisms: prey reliability, prey biomass, and prey composition mechanisms. The results suggest that greater prey reliability at higher prey diversities enhances predator stability and provide support for MacArthur (1955).


Introduction

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

Theory shows that increases in food-web complexity can increase (MacArthur 1955; Elton 1958; DeAngelis 1975) or decrease (May 1973; Pimm 1982) stability. The relation between complexity and stability depends on how food-web complexity is manipulated, the level of organization at which stability is measured (McNaughton 1977; King & Pimm 1983), and the meaning of stability (Lewontin 1969). Stability (measured as temporal variability) often decreases with increasing food-web complexity when population stability is measured, and often increases with complexity when the stability of an aggregate, such as total biomass of all species, is measured (Tilman 1996; McGrady-Steed & Morin 2000). This paper concerns how aspects of predator population stability are affected by prey species diversity. This is interesting because the stability of a predator population may be determined by the stability of an aggregate – the sum of all prey species. I used persistence time, population size (where larger populations are less likely to experience zero density by chance), and population variability as three practical measures of population stability (Lewontin 1969). MacArthur (1955) put forward a hypothesis of how higher prey diversity could increase predator stability; I now present a thorough treatment of the possible mechanisms.

Mechanisms that cause higher predator stability when more prey species are available fall into three categories (Table 1 includes references).

Table 1.  Three groups of mechanisms can cause increases in predator stability when more prey species are available for consumption. Italic text shows the specific predictions tested in this study
 MechanismReferencesSpecific prediction
(I) Prey reliability mechanisms
(Resource reliability increases with resource diversity and decreases predator population variability)(a) Smaller effect of losing one prey species on total prey biomassMacArthur (1955)Negative relation between relative effect of losing one prey species on predator density and prey diversity
 (b) Lower temporal variability in total prey biomassMcNaughton (1977); King & Pimm (1983); Doak et al. (1998); Tilman et al. (1998); Yachi & Loreau (1999); Ives et al. (1999)Negative relation between prey variability and prey diversity; positive relation between predator variability and prey variability
(II) Prey biomass mechanisms
(Total prey biomass increases with prey diversity and increases predator population size)(a) Sampling effect of prey diversityAarssen (1997); Huston (1997); Tilman, Lehman & Thomson (1997b)Prey biomass is mostly explained by presence of particular prey species
 (b) More complete resource use by preyBolker et al. (1995); Chapin et al. (1997); Tilman et al. (1997b); Loreau (1998)Over-yielding by prey
 (c) Greater chance/frequency of facilitative interactionsHooper (1998); Hooper & Vitousek (1997); van der Heijden et al. (1998)Over-yielding by prey
(III) Prey composition mechanisms
(Presence of particular combinations of prey species determine predator density)(a) Complementary nutritional contents of prey speciesPulliam (1975); Belovsky (1978); Westoby (1978)Higher predator density than expected given available prey biomass and observed predator density on single prey species
(b) Reduction in resource uptake rate by addition of a poor quality resource speciesAbrams (1993); Kretzschmar et al. (1993)Lower predator density than expected given available prey biomass and observed predator density on single prey species

(I) Prey reliability mechanisms: greater prey diversity can increase the reliability (stability) of the predators aggregate resource pool, i.e. not all prey species become rare simultaneously, and should therefore decrease predator population variability. This could happen through various mechanisms: (Ia) essentially MacArthur's hypothesis, in an environment with a variable number of prey, the more prey that are present, the smaller the effect of losing an individual prey species; (Ib) temporal variation of aggregates, such as total prey biomass, tend to decrease as the number of components (prey species) making up the aggregate increases.

(II) Prey biomass mechanisms: increases in prey diversity lead to increases in total prey biomass. Higher prey biomass leads to higher predator abundance, and higher predator abundance reduces the risk of extinctions by demographic and environmental stochasticity (Lande 1993). This hypothesis requires a mechanism whereby increased prey diversity leads to greater total prey biomass. Examples are: (IIa) sampling effects of prey diversity, where the presence of a prey species with particularly high biomass is more likely in more diverse prey assemblages (this is identical to the sampling effect of diversity on ecosystem function); (IIb) more complete use of available resources by diverse prey assemblages when prey species use different resources; (IIc) greater chance/frequency of facilitative interactions.

(III) Prey composition mechanisms where particular combinations of prey species cause greater predator abundance and therefore stability. Examples are: (IIIa) when a predator benefits from eating a diverse range of prey species because single prey species do not contain all the nutrients necessary for growth and reproduction; (IIIb) in contrast, a lower consumption rate of a nutritious prey species due to the presence of a palatable but less nutritious prey species might decrease the overall rate of resource intake by the predator (this would cause lower predator density at higher levels of prey diversity).

A variety of studies provide evidence for some of these mechanisms. Observational studies (Pimm 1991) provide tentative evidence for prey reliability mechanisms because some populations of polyphagous insect herbivores are less variable than monophagous insect herbivores (Watt & Craig 1986; Redfearn & Pimm 1988; but see Owen & Gilbert 1989). Other ecological studies supply a consumer with different numbers of prey species (Hairston et al. 1968; Luckinbill 1979; Lawler & Morin 1993; Karban, Hougen-Eitzmann & English-Loeb 1994; Balčiūnas & Lawler 1995; Bonsall & Hassell 1997) and often show greater consumer abundance when fed on more prey species (evidence for prey biomass or prey composition mechanisms). Neither approach, however, allows explicit tests of specific mechanisms. A recent study of acorn woodpeckers (Melanerpes formicivorus Swainson) assessed the importance of resource (acorns) reliability and total resource abundance on woodpecker variability and density (Koenig & Haydock 1999). Geographic areas with more species of oak had more reliable acorn supplies and less variable woodpecker populations. In addition, areas with greater total oak abundance contained higher woodpecker density. These results suggest that prey reliability mechanisms and prey abundance/biomass mechanisms affect predator variability and abundance, respectively.

Here I document a positive relation between prey diversity and population stability of the predatory ciliate Dileptus anser (Müller) and explore which of the mechanisms in Table 1 could contribute to this relation. I fed Dileptus one, two, or three prey species to establish the relationship between predator stability and prey diversity. Because multi-generation time series of population dynamics are needed to estimate population stability, and because of the relative ease of manipulation and replication, microbial microcosms were ideal model communities for this study (Lawton 1995; Lawler 1998). Dileptus is an appropriate predator because it consumes a wide variety of prey species, and direct laboratory observations of feeding behaviour suggest little preference for different prey species. Therefore, manipulating the number of available prey species directly controls the number of prey species Dileptus consumes. Because species composition is an important determinant of community properties (Lawler 1993; Hooper & Vitousek 1997; Huston 1997; McGrady-Steed, Harris & Morin 1997; Tilman et al. 1997a; Hooper 1998), and inevitably changes along a diversity gradient, I used different prey compositions within two of the diversity levels.

Predator population persistence time, population density, and population variability were measured. Each is an important measure of stability because each is closely related to extinction probability. Persistence time records an extinction event, population density is negatively correlated with extinction probability and population variability is positively correlated with extinction probability (Lande 1993). Predator dynamics and prey dynamics with and without Dileptus allowed me to test for prey reliability mechanisms and prey composition mechanisms.

Materials and methods

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

Protist microcosms

Culture methods closely followed Lawler & Morin (1993). Microcosms were 240 mL covered glass bottles placed in a 16 h: 8 h light: dark photoperiod incubator at 22 °C. Each bottle contained 100 mL of nutrient medium (0·55 g Carolina Biological Supply protozoan pellet per litre of well water) and two sterile wheat seeds to provide additional nutrients. Inoculation of the sterile medium with bacteria (Bacillus cereus, Bacillus subtilus and Serratia marcescens) before the medium was divided into microcosms standardized initial bacterial conditions. Bacterivorous protists formed the second trophic level (the prey). The three bacterivores were the flagellate Collodictyon triciliatum (Carter) (L = mean length ± 1 SE = 67 ± 3 μm; W = mean width ± 1 SE = 45 ± 2 μm; M = mean cell mass ± 1 SE = 1·2 ± 0·1 × 10−6 mg; n = 10), and the ciliates Colpidium striatum (Ehrenberg) (L = 162 ± 14 μm; W = 75 ± 14 μm; M = 12 ± 1 × 10−6 mg; n = 10) and Paramecium aurelia (Ehrenberg) (L = 228 ± 6 μm; W = 66 ± 3 μm; M = 130 ± 11 × 10−6 mg; n = 10). The large predatory ciliate Dileptus anser (L including proboscis = 500 ± 30 μm; W = 67 ± 0 μm; M = 40 ± 3 × 10−3 mg, n = 6) was added to form a third trophic level. Colpidium, Paramecium and Dileptus were obtained from Carolina Biological Supply, Burlington, USA; Collodictyon is an occasional contaminant of stock cultures that I isolated for this study. The bacterivores were introduced into microcosms by adding five pipette drops of stock culture media of the appropriate species. Twenty rinsed Dileptus individuals were added to microcosms 7 days later after bacterivores attained high densities. All measures of time are presented as days relative to the day (zero) when Dileptus was added to the microcosms. Microcosms were inoculated with bacteria on day −8, bacterivores on day −7 and Dileptus on day 0.

Experimental design

All combinations of the three bacterivore prey species Colpidium (C) Collodictyon (F for flagellate) and Paramecium (P) made seven levels (C, F, P, CF, CP, FP and CFP) of the prey composition treatment. The prey composition levels also generated differences in prey diversity from one to three prey species. Each treatment combination was replicated five times, for a total of 35 microcosms.

I estimated population density of the three prey species and Dileptus by withdrawing a sample of known weight (c. 0·3 mL) from each microcosm and counting the number of protists under a Nikon SMZ-U dissecting microscope. Samples were diluted by weight if densities were too high to count. Microcosms were sampled every 2 days starting on day 2 and ending on day 14–20. Extinction of Dileptus was confirmed by shining a bright light through the microcosms and directly searching for Dileptus, which are large enough to be seem with the naked eye.

Some of the mechanisms in Table 1 can be tested with information about prey dynamics in the absence of a predator. I assembled five replicates of each prey composition (totalling 35 microcosms) without Dileptus and estimated population density every 3–4 days. All other details were as in the previous experiment.

Data analysis

Persistence time, mean population density (time average of log(density mL−1 + 1)) and coefficient of variation of untransformed population density (McArdle, Gaston & Lawton 1990) provided different measures of predator population dynamics. Persistence times were censored (taken as the length of the experiment) when the experiment was terminated before a population went extinct. The initial growth phase of Dileptus populations was excluded from the calculations of population density and variability. Linear regression tested the relationship between each predator population dynamic measure and prey species number with prey species compositions as replicates (n = 7). Because of possible heterogeneous variance structure I also used a randomization test (Manly 1997) to calculate the probability of significant relations between prey diversity and each measure of predator stability. The randomization test compared the observed slope to the distribution of slopes given from all possible permutations of the data. Differences in predator population dynamics caused by prey composition were evaluated with one-way anova for each of the stability measures (variances were not significantly heterogeneous by Barlett's test). Prey density in the absence of Dileptus was time averaged log(density mL−1 + 1) excluding the initial growth phase.

Results

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

Prey diversity and predator stability

Dileptus showed a range of dynamics (Fig. 1), from low-density populations (F and P), to boom–bust patterns (C, CF, CP, and FP), to high-density persistent populations (CFP). Population variability decreased significantly with prey species number (Fig. 2c), suggesting that the strongest stabilizing effect of prey diversity operated through increases in prey reliability. Persistence time and population density both increased with prey species number but the relations were not significant at α= 0·05 (Fig. 2a,b).

image

Figure 1. Example of predator–prey population dynamics for each prey species combination.

Download figure to PowerPoint

image

Figure 2. Relations between prey diversity and (a) predator persistence time, (b) population density, and (c) population variability. Linear regression lines using composition means as replicates are shown, as well as the regression equation, r2, standard P-value, and P-value from the randomization test (see text).

Download figure to PowerPoint

Prey composition significantly affected each measure of predator population dynamics (Fig. 3). For example, Dileptus feeding on Colpidium alone attained greater density than when feeding on a mixture of Colpidium and Paramecium, and Dileptus grew to similar densities when feeding on two-prey mixtures (CF & FP) and the three-prey mixture (CFP). Although predator population stability increased with prey species richness, the pattern would not necessarily hold if only particular mixtures of prey species were included in the analysis.

image

Figure 3. Prey composition affected predator persistence time, population density, and population variability (error bars are ± 1 SE). Persistence time anovaF6,20 = 5·0, P < 0·01, r2 = 0·60; density anovaF6,20 = 20·9, P < 0·0001, r2 = 0·86; variability anovaF6,20 = 2·7, P < 0·05, r2 = 0·45. Replication is shown in the plot of population variability and is the same for all plots (Dileptus failed to establish in one F, CF, CP and CFP replicate and two P and FP replicates). Parenthesized numbers beside persistence times show the number of censored persistence times used in calculations.

Download figure to PowerPoint

Testing for the mechanisms

(Ia) I estimated the relative effect of losing a prey species by comparing predator densities in different treatments. For example, the effect of losing Colpidium from the three-prey species combination, was estimated as [predator density on CFP – predator density on FP]/predator density on CFP. This effect tended to decrease with increasing prey diversity (Fig. 4). Inspecting the time series (Fig. 1) within treatments also suggests that the effect of losing a prey species is, at least in some cases, smaller when more prey remain. Dileptus drives Colpidium extinct or to very low density whenever it is present, and Dileptus then decreases quickly to extinction when there are no other prey available (Fig. 1, composition C), but remains at high density when other prey species are available (Fig. 1, composition CFP).

image

Figure 4. The relative effect of losing a prey on predator density, calculated from mean predator densities on different prey compositions. For example Colpidium, [predator density on CFP – predator density on FP]/predator density on CFP (therefore positive numbers indicate loss of a prey species resulted in lower predator density). Losing a prey species caused a reduction in predator density in most cases, but greatly increased predator density in one case. Statistics are not informative because data points are not independent.

Download figure to PowerPoint

(Ib) Variability in total prey biomass should decrease with increasing prey diversity. There was no significant negative relation between variability (CV) in total prey biomass and prey diversity with (linear regression, d.f. = 1,5, y = 77 + 9x;P = 0·77) or without (linear regression, d.f. = 1,5, y = 54 + 0·3x;P = 0·97) Dileptus present.

(II) & (III) Effects of prey biomass or effects of particular combinations of prey species might determine predator density. There was evidence that Dileptus experienced nutritional constraints (IIIa) when feeding on only Collodictyon or Paramecium that were released when feeding on both Collodictyon and Paramecium. Dileptus grew poorly on Collodictyon alone (mean density = 0·6 mL−1) or Paramecium alone (mean density = 0·56 mL−1) and should have reached c. 1·16 mL−1 (0·6 + 0·56) in the presence of both prey species if they acted additively. Dileptus density was, however, much higher (12·5 mL−1) when both prey species were present than expected by additivity. The extra Dileptus growth with both present cannot be explained by extra prey biomass because the total biomass of Collodictyon and Paramecium grown together (17 × 10−3 mg mL−1) was less than the sum of both when grown alone (29 × 10−3 mg mL −1). Additional analyses using prey-specific predator numerical responses and prey biomass to predict Dileptus density support this conclusion but were not sensitive enough to detect experiment wide effects of combinations of prey species (unpublished data).

There was also evidence for the sampling effect of prey diversity (IIa). The presence of Colpidium was coincident with and responsible for high prey biomass just before the addition of Dileptus (Fig. 5). High prey biomass is more likely at high levels of prey diversities because Colpidium is more likely to be present. However, this sampling effect cannot be linked directly to predator density because prey composition mechanisms (III) can affect predator density.

image

Figure 5. Total prey density in each of the seven prey compositions just before predator addition and contribution of each prey species to the total. Error bars are ± 1 SE for each species in a column and n = 5 for all stacked bars. Letters directly above error bars show Tukey's groupings with α= 0·05 and coincide with the presence or absence of Colpidium.

Download figure to PowerPoint

Discussion

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

This study documents the positive effect of prey diversity on predator stability. This diversity–stability relation contrasts with many studies linking greater food-web complexity to lower population stability (May 1973; Pimm 1991; Tilman 1996; McGrady-Steed & Morin 2000). Effects of prey diversity on predator stability in this study (see also Hairston et al. 1968; Koenig & Haydock 1999) describe a situation very similar to that envisaged by MacArthur (1955), where more pathways (prey species) for energy acquisition stabilize the consumer population dependent on that energy (the predator population). This is very different from a situation where increased diversity is confounded with food chain length, or is distributed over several trophic levels. The number of trophic levels, overall number of species, and number of prey species should affect population stability in different ways because effects of complexity on population stability depend on what aspect of complexity is considered even when the same definition of stability is employed (Goodman 1975; McNaughton 1977; King & Pimm 1983; Pimm 1984, 1991; McGrady-Steed & Morin 2000).

In general, several mechanisms can contribute to the effects of prey diversity on predator stability (Table 1). The significant relation between predator population variability and prey diversity in this study (Fig. 2c) suggests that prey reliability mechanisms, such as MacArthur's (1955), can cause increased predator stability, but does not of course rule out other mechanisms. In addition, predator density was less affected by loss of species when other species remained, providing further support for MacArthur's hypothesis. The data ruled out lower temporal variability in total prey biomass (Ib) as a mechanism contributing to the positive relation between prey diversity and predator stability. There was no decrease in prey temporal variability with increased prey diversity; rather, the trend was towards higher variability, contrary to some theoretical expectations (McNaughton 1977; King & Pimm 1983; Doak et al. 1998; Yachi & Loreau 1999). Greater prey diversity can decrease total prey variability because of statistical averaging (Doak et al. 1998 but see Tilman, Lehman & Bristow 1998) and also when fluctuations of species are negatively correlated, a possible outcome of competitive interactions (Doak et al. 1998; Yachi & Loreau 1999). These stabilizing effects can be cancelled if competitive interactions increase population variability of competitors (the prey here) (Ives, Gross & Klug 1999). Greater Colpidium population variability at higher levels of prey diversity (mean CV without Dileptus; alone = 68, with one competitor = 90, and with two competitors = 132) supports destabilizing effects of competitive interactions on population dynamics and explains why variability of total prey biomass tended to increase with prey diversity. (Collodictyon and Paramecium population variability changed idiosyncratically with prey diversity so total prey variability was largely determined by Colpidium population variability.)

All the remaining mechanisms in Table 1 could affect predator population dynamics and contribute to relations between prey diversity and predator stability. Notably, there was good evidence that the biomass of different prey species did not always act additively, so predator density was not a simple function of total prey biomass. Further experiments using more prey compositions within diversity levels, a greater range of prey species richness, and methods to distinguish between prey biomass mechanisms and prey composition mechanisms would allow a more sensitive and complete examination of the determinants of predator population dynamics.

Although the average effect of increasing prey diversity was to increase predator stability, differences in prey species composition strongly influenced predator stability. This result reinforces existing evidence that species composition is an important determinant of pattern in ecological systems (Tilman, Wedin & Knops 1996; Hooper & Vitousek 1997; Hooper 1998). There is, however, no conflict between the effects of composition and number of species. The effect of prey species number is a result of the distribution of species composition effects over the range of prey species richness. The importance of species composition, as well species diversity underscores the need to separate effects of species composition and diversity (Givnish 1994; Huston 1997). Use of different species compositions within levels of diversity allow the relative importance of composition and diversity effects to be assessed. With only one species composition within the three prey species diversity level, I cannot completely separate effects of composition and diversity, and further compositions and diversity levels would again be useful.

Concordant increases in predator stability and prey diversity have important implications. Higher prey biomass caused by greater productivity can destabilize predator–prey interactions through the paradox of enrichment (Rosenzweig 1971; Luckinbill 1974, 1979). However, increases in prey biomass caused by increasing prey diversity are clearly different from increases in prey biomass caused by enrichment. Differences between prey species may stabilize the interaction in some way (McCauley & Murdoch 1990), perhaps through differences in prey vulnerability (Abrams 1993; Kretzschmar, Nisbet & McCauley 1993; Abrams & Walters 1996). Because greater prey diversity increases the stability of a further trophic level, high prey diversity could contribute towards the maintenance of long food chains (Hutchinson 1959; Pimm & Lawton 1977; Oksanen et al. 1981; Cohen, Briand & Newman 1990). More prey species can also stabilize a community by increasing the number of self-regulated populations (Sterner, Bajpai & Adams 1997), adding another hypothesis how greater prey diversity can maintain long food chains.

Acknowledgements

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

Peter Morin motivated this work and made many valuable comments during the work and on earlier drafts. Thanks to all members of the Morin Lab, especially Jeremy Fox, Christina Kaunzinger, Jill McGrady-Steed, and Henry Stevens, for their help. Funding came from Cook College, Rutgers University, The Biodiversity Center at Cook College, and NSF grant 9806427 to Peter Morin.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Aarssen, L.W. (1997) High productivity in grassland ecosystems: effected by species diversity or productive species. Oikos, 80, 183 184.
  • Abrams, P.A. (1993) Effect of increased productivity on the abundances of trophic levels. American Naturalist, 141, 351 371.
  • Abrams, P.A. & Walters, C.J. (1996) Invulnerable prey and the paradox of enrichment. Ecology, 77, 1125 1133.
  • Balčiūnas, D. & Lawler, S.P. (1995) Effects of basal resources, predation, and alternate prey in microcosm food chains. Ecology, 76, 1327 1336.
  • Belovsky, G.E. (1978) Diet optimization in a generalist herbivore: the moose. Theoretical Population Biology, 14, 105 134.
  • Bolker, B.M., Pacala, S.W., Bazzaz, F.A., Canham, C.D., Levin, S.A. (1995) Species diversity and ecosystem response to carbon dioxide fertilization: conclusions from a temperate forest model. Global Change Biology, 1, 373 381.
  • Bonsall, M.B. & Hassell, M.P. (1997) Apparent competition structures ecological assemblages. Nature, 388, 371 373.
  • Chapin, F.S., Walker, B.H., Hobbs, R.J., Hooper, D.U., Lawton, J.H., Sala, O.E., Tilman, D. (1997) Biotic control over the functioning of ecosystems. Science, 277, 500 504.
  • Cohen, J.E., Briand, F., Newman, C.M. (1990). Community Food Webs. Springer-Verlag. Berlin.
  • DeAngelis, D.L. (1975) Stability and connectance in food web models. Ecology, 56, 238 243.
  • Doak, D.F., Bigger, D., Harding, E.K., Marvier, M.A., O'Malley, R.E., Thomson, D. (1998) The statistical inevitability of stability–diversity relationships in community ecology. American Naturalist, 151, 264 276.
  • Elton, C.S. (1958) The Ecology of Invasions by Animals and Plants. Chapman & Hall, London.
  • Givnish, T.J. (1994) Does diversity beget stability? Nature, 371, 113 114.
  • Goodman, D. (1975) The theory of diversity–stability relationships in ecology. Quarterly Review of Biology, 50, 237 266.
  • Hairston, N.G., Allan, J.D., Colwell, R.K., Futuyma, D.J., Howell, J., Lubin, M.D., Mathias, J., Vandermeer, J.H. (1968) The relationship between species diversity and stability: an experimental approach with protozoa and bacteria. Ecology, 49, 1091 1101.
  • Van Der Heijden, M.G.A., Klironomos, J.N., Ursic, M., Moutoglis, P., Streitwolf-Engel, R., Boller, T., Wiemken, A., Sanders, I.R. (1998) Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature, 396, 69 72.
  • Hooper, D.U. (1998) The role of complementarity and competition in ecosystem responses to variation in plant diversity. Ecology, 79, 704 719.
  • Hooper, D.U. & Vitousek, P.M. (1997) The effects of plant composition and diversity on ecosystem processes. Science, 277, 1302 1305.
  • Huston, M.A. (1997) Hidden treatments in ecological experiments: re-evaluating the ecosystem function of biodiversity. Oecologia, 110, 449 460.
  • Hutchinson, G.E. (1959) Homage to Santa Rosalia, or why are there so many kinds of animals? American Naturalist, 93, 145 159.
  • Ives, A.R., Gross, K., Klug, J.L. (1999) Stability and variability in competitive communities. Science, 286, 542 544.
  • Karban, R., Hougen-Eitzmann, D., English-Loeb, G. (1994) Predator-mediated apparent competition between two herbivores that feed on grapevines. Oecologia, 97, 508 511.
  • King, A.W. & Pimm, S.L. (1983) Complexity, diversity, and stability: a reconciliation of theoretical and empirical results. American Naturalist, 122, 229 239.
  • Koenig, W.D. & Haydock, J. (1999) Oaks, acorns, and the geographical ecology of acorn woodpeckers. Journal of Biogeography, 26, 159 165.
  • Kretzschmar, M., Nisbet, R.M., McCauley, E. (1993) A predator–prey model for zooplankton grazing on competing algal populations. Theoretical Population Biology, 44, 32 66.
  • Lande, R. (1993) Risks of population extinction from demographic and environmental stochasticity and random catastrophes. American Naturalist, 142, 911 922.
  • Lawler, S.P. (1993) Species richness, species composition and population dynamics of protists in experimental microcosms. Journal of Animal Ecology, 62, 711 719.
  • Lawler, S.P. (1998) Ecology in a bottle: using microcosms to test theory. Experimental Ecology. Issues and Perspectives (eds W.J.J.Resetarits & J.Bernardo), pp. 236 253. Oxford University Press, Oxford.
  • Lawler, S.P. & Morin, P.J. (1993) Food web architecture and population dynamics in laboratory microcosms of protists. American Naturalist, 141, 675 686.
  • Lawton, J.H. (1995) Ecological experiments with model systems. Science, 269, 328 331.
  • Lewontin, R.C. (1969) The meaning of stability. Diversity and Stability in Ecological Systems (eds G.M.Woodwell & H.H.Smith), pp. 13 23. National Technical Information service, Brookhaven National Laboratory.
  • Loreau, M. (1998) Biodiversity and ecosystem functioning: a mechanistic model. Proceedings of the National Academy of Sciences USA, 95, 5632 5636.
  • Luckinbill, L.S. (1974) The effects of space and enrichment on a predator–prey system. Ecology, 55, 1142 1147.
  • Luckinbill, L.S. (1979) Regulation, stability, and diversity in a model experimental microcosm. Ecology, 60, 1098 1102.
  • MacArthur, R. (1955) Fluctuations of animal populations, and a measure of community stability. Ecology, 36, 533 536.
  • Manly, B.F.J. (1997). Randomization, Bootstrap and Monte Carlo Methods in Biology. Chapman & Hall, London.
  • May, R.M. (1973). Stability and Complexity in Model Ecosystems. Princeton University Press, Princeton.
  • McArdle, B.H., Gaston, K.J., Lawton, J.H. (1990) Variation in the size of animal populations: patterns, problems and artefacts. Journal of Animal Ecology, 59, 439 454.
  • McCauley, E. & Murdoch, W.W. (1990) Predator–prey dynamics in environments rich and poor in nutrients. Nature, 343, 455 457.
  • McGrady-Steed, J., Harris, P.M., Morin, P.J. (1997) Biodiversity regulates ecosystem predictability. Nature, 390, 162 165.
  • McGrady-Steed, J. & Morin, P.J. (2000) Biodiversity, density compensation, and the dynamics of populations and functional groups. Ecology, 81, 361 373.
  • McNaughton, S.J. (1977) Diversity and stability of ecological communities: a comment on the role of empiricism in ecology. American Naturalist, 111, 515 525.
  • Oksanen, L., Fretwell, S.D., Arruda, J., Niemelä, P. (1981) Exploitation ecosystems in gradients of primary productivity. American Naturalist, 118, 240 261.
  • Owen, J. & Gilbert, F.S. (1989) On the abundance of hoverflies (Syrphidae). Oikos, 55, 183 193.
  • Pimm, S.L. (1982). Food Webs. Chapman & Hall, London.
  • Pimm, S.L. (1984) The complexity and stability of ecosystems. Nature, 307, 321 326.
  • Pimm, S.L. (1991). The Balance of Nature? University of Chicago Press, Chicago.
  • Pimm, S.L. & Lawton, J.H. (1977) Number of trophic levels in ecological communities. Nature, 268, 331.
  • Pulliam, H.R. (1975) Diet optimization with nutrient constraints. American Naturalist, 109, 765 768.
  • Redfearn, A. & Pimm, S.L. (1988) Population variability and polyphagy in herbivorous insect communities. Ecological Monographs, 58, 39 55.
  • Rosenzweig, M.L. (1971) Paradox of enrichment: destabilization of exploitation ecosystems in ecological time. Science, 171, 385 387.
  • Sterner, R.W., Bajpai, A., Adams, T. (1997) The enigma of food chain length: absence of theoretical evidence for dynamic constraints. Ecology, 78, 2258 2262.
  • Tilman, D. (1996) Biodiversity: population versus ecosystem stability. Ecology, 77, 350 363.
  • Tilman, D., Knops, J., Wedin, D., Reich, P., Ritchie, M., Siemann, E. (1997a) The influence of functional diversity and composition on ecosystem processes. Science, 277, 1300 1302.
  • Tilman, D., Lehman, C.L., Bristow, C.E. (1998) Diversity–stability relationships: statistical inevitability or ecological consequence? American Naturalist, 151, 277 282.
  • Tilman, D., Lehman, C.L., Thomson, K.T. (1997b) Plant diversity and ecosystem productivity: theoretical considerations. Proceedings of the National Academy of Sciences USA, 94, 1857 1861.
  • Tilman, D., Wedin, D., Knops, J. (1996) Productivity and sustainability influenced by biodiversity in grassland ecosystems. Nature, 379, 718 720.
  • Watt, K.E.F. & Craig, P.P. (1986) System stability principles. Systems Research, 3, 191 201.
  • Westoby, M. (1978) What are the biological bases of varied diets? American Naturalist, 112, 627 631.
  • Yachi, S. & Loreau, M. (1999) Biodiversity and ecosystem productivity in a fluctuating environment: the insurance hypothesis. Proceedings of the National Academy of Sciences USA, 96, 1463 1468.

Received 10 May 1999;revisionreceived 29 March 2000