Freshwater Biology (2005) 50, 105–112 doi:10.1111/j.1365-2427.2004.01310.

Temporal stability of pond zooplankton assemblages

CHRISTOPHER F. STEINER
W. K. Kellogg Biological Station and the Department of Zoology, Michigan State University, Hickory Corners, MI, U.S.A.

SUMMARY
1. A large body of recent theory has recently developed focused on the relationship
between the species diversity of competitor assemblages and the temporal stability of total
competitor biomass. Many of these models predict that stability can increase with
increasing diversity.
2. To explore natural relationships between zooplankton taxonomic diversity and temporal
stability of total zooplankton biomass, 18 fishless, permanent ponds located in southern
Michigan were surveyed over a 5 month period during a single growing season.
3. Results showed that temporal variability in total zooplankton biomass (measured as the
coefficient of variation or CV) decreased with increasing mean zooplankton taxonomic
richness. Thus, temporal stability increased with increasing taxonomic richness, consistent
with theoretical predictions.
4. Decreases in the CV appeared to be because of portfolio effects (statistical averaging
of species’ biomass fluctuations) rather than negative covariances among zooplankton
taxa.
5. The CV of zooplankton biomass was also related to several environmental variables,
suggesting that taxonomic richness may not be the only mediator of biomass stability. The
CV decreased with increasing relative abundance of grazer-resistant algae (algae >35 lm
in size) and the CV increased with increasing pond productivity.

Keywords: biomass, ecosystem functioning, species diversity, stability, zooplankton

linked to valuable ecosystem services, such as carbon-

Introduction
The relationship between species diversity and system
stability is a central issue in ecological research and
has recently gained renewed attention in light of
widespread, human-induced alterations of global
biodiversity (Tilman, 1999; McCann, 2000; Loreau
et al., 2002). An essential aspect of community and
ecosystem stability is the variability through time of
aggregate community measures such as the total
biomass of trophic levels or competitive assemblages
(Tilman, 1999). Comprehending the impact of species
loss on system stability is vital for predicting the
reliability of such aggregate measures through time.
This is especially true when aggregate measures are
Correspondence: Christopher F. Steiner, Department of Animal
Biology, University of Illinois, 515 Morrill Hall,
505 South Goodwin Avenue, Urbana, IL 61801, U.S.A.
E-mail: csteiner@life.uiuc.edu
nutrient cycling or the biomasses of harvestable and
managed biota. Theory provides a rich foundation of
mechanisms and predictions for the impact of species
diversity on community stability (reviewed in Tilman,
1999; Cottingham, Brown & Lennon, 2001; Hughes,
Ives & Norberg, 2002; Loreau et al., 2002). Despite this,
documentation of diversity-stability patterns in nat-
ural systems is rare.
In this paper, I focus on the relationship between
temporal variability in total competitor biomass and
the richness of competing species (i.e. diversity based
on presence and absence of taxa). ‘Competitor’ is used
to refer to species that occupy a single trophic level
(e.g. herbivorous zooplankton) and potentially com-
pete for shared resources (e.g. algae). Theoretical
explorations show that increasing species richness of
competitors can enhance the temporal stability of total
competitor biomass through a number of mechanisms
(Doak et al., 1998; Tilman, Lehman & Bristow, 1998;

2004 Blackwell Publishing Ltd 105

106 C.F. Steiner
Ives, Gross & Klug, 1999; Yachi & Loreau, 1999; Ives,
Klug & Gross, 2000; Ives & Hughes, 2002; reviewed in
Tilman, 1999; Cottingham et al., 2001; Hughes et al.,
2002; Loreau et al., 2002). Consider a common meas-
ure of temporal variability, the coefficient of variation
(CV), equal to the standard deviation (rt) of the
variable of interest measured through time divided by
the temporal mean (lt) of that variable (CV ¼ rt/lt).
The standard deviation of competitor biomass is itself
composed of the sum of all competitor species
variances and covariances in biomass measured over
time: rt ¼ R variance + (2 ·R covariance)1/2. Hence,
biomass stability will depend on species richness if
either variance or covariance components vary with
species richness. Summed variances can decline with
increasing species richness via statistical averaging,
enhancing stability in the complete absence of species
interactions; what is termed the portfolio effect (Doak
et al., 1998; Tilman et al., 1998; see Discussion). How-
ever, species interactions may also influence rt by
affecting how species’ biomasses covary through time.
For example, competitive effects may lead to negative
covariances between some species. Alternatively,
species within a trophic-level may respond differen-
tially to environmental fluctuations and perturbations
(Ives et al., 1999; Yachi & Loreau, 1999; Ives et al.,
2000). If of sufficient number and/or magnitude, the
sum of all covariances (R covariance) may be negative,
decreasing temporal variability in total competitor
biomass. If summed covariances become more negat-
ive with increasing species richness, stability may
increase with diversity (Tilman, 1999).
Although temporal stability of total competitor
biomass may increase with increasing species rich-
ness, numerous other factors may independently
influence the stability of populations and communi-
ties. For instance, system productivity and enrichment
have long been thought to affect negatively the
stability of populations. Simple consumer-resource
models show that such systems are more prone to
exhibit unstable-oscillatory dynamics under enriched
conditions; what is commonly known as the paradox
of enrichment (Rosenzweig, 1971; Gilpin, 1972;
McCauley & Murdoch, 1990). How such consumer-
prey dynamics scale up to whole consumer assem-
blages and total consumer biomass is unknown. More
recently, theoretical and empirical research has shown
that increasing prey diversity and inclusion of
inedible or consumer-resistant prey species (or weak
interactors) can stabilise consumer–prey dynamics
(Abrams & Walters, 1996; McCann, Hastings & Huxel,
1998; Bohannan & Lenski, 1999; McCauley et al., 1999;
Petchey, 2000). Thus, the prevalence of grazer-resist-
ant algal-prey could stabilise zooplankton dynamics
(e.g. McCauley et al., 1999). These investigations have
not considered complex consumer assemblages or
effects on community-level biomass.
Studies that have used survey/observational data
to examine natural relationships between species
richness and variability in trophic level biomass are
surprisingly rare. Here I present results from a field
survey of pond zooplankton communities in southern
Michigan. Using this data set, I explore the relation-
ship between mean zooplankton taxonomic richness
and temporal stability of total zooplankton biomass. I
also examine several other limnological variables that
may have affected these stability measures.
Methods
All ponds were found within 100 km of the W. K.
Kellogg Biological Station (Hickory Corners, MI,
U.S.A.), were fishless, and permanent in the 1–2 years
prior to the study (personal observation). Ponds were
sampled monthly in 1998, beginning in late April and
ending in late August. For methodological details see
Steiner (2002) and Steiner (2004). In most cases,
zooplankton were collected at mid-day with integra-
ted tube samplers that extended to the pond bottom,
with the exception of ponds approximately <0.5 m in
depth (which were sampled with a 2 L hand pitcher).
Samples were collected at several points, at even
intervals, along a qualitative transect spanning from
approximately 1 m in the pond from edge to the
centre of each pond. At each sample point along the
transect, the water column was sampled and the
water poured through a 60 lm sieve to retain
zooplankton; this was then repeated and a sample of
equal volume poured through a separate 60 lm sieve.
Thus, two replicate zooplankton samples were taken,
each integrating spatial variation in the pond. Total
sample volume varied among ponds and ranged
between 24 and 108 L. Larger sample volumes
obtained in a subset of the study ponds spanning a
range of productivities revealed that sample volumes
of 10–24 L were adequate for detecting the majority
(>80%) of zooplankton taxa. Zooplankton were pre-
served in acid Lugol’s solution and later identified to

2004 Blackwell Publishing Ltd, Freshwater Biology, 50, 105–112

genus or species, excepting copepods, which were
classed as calanoid or cyclopoid. Rotifers were iden-
tified to genus or species with the exception of
members of the Notommatidae and Proalidae, which
were identified to family, and the Bdelloidea, which
were identified to the level of Order. Protozoa and
copepod nauplii were not counted. In each sample, up
to 50 randomly chosen individuals of each taxon were
measured to obtain dry mass estimates using pub-
lished length-mass regressions (McCauley, 1984).
Because of low water levels, zooplankton could not
be sampled in eight ponds in August.
In addition to zooplankton, several additional biotic
and abiotic variables were measured monthly in the
ponds. Oxygen concentration, pH and temperature
were measured along transects using a Horiba U-10
multi-probe (Horiba Ltd., Kyoto, Japan). Pond water
(1.5 L total) was collected at each sample point along
the transect using the same samplers used to collect
zooplankton, pooled and immediately placed on ice,
in the dark, for later analysis of total phosphorus, total
chlorophyll a, chlorophyll a of algae <35 lm in size,
and the C : P and N : P content of seston less than
35 lm in size (see Steiner, 2002, 2004). To size
fractionate samples, water was first filtered through
a 35 lm mesh. I used 35 lm as a size-based cutoff
between highly edible algae and grazer-resistant
algae. Although several algal traits may affect edib-
ility, size is known to be an important feeding
constraint (Sterner, 1989) and 35 lm has been used
previously as an effective measure of grazer suscep-
tibility (e.g. Carpenter et al., 1987; Sarnelle, 1992;
Cottingham, 1999; Steiner, 2003a). Chlorophyll a and
seston C : N : P data are discussed in Steiner (2002).
Means and ranges of several pond environmental
variables, in addition to analyses of zooplankton
composition and patterns of Daphnia dominance, can
be found in Steiner (2004).
I used estimates of zooplankton dry mass as a
measure of biomass. For each sample date, replicate
measures of total zooplankton biomass were first
averaged to obtain a single measure. Temporal
biomass variability in each pond was then measured
as the coefficient of variation (CV) through time of
total zooplankton biomass. To measure zooplankton
taxonomic richness for each pond, total taxonomic
richness among both replicate zooplankton samples
was first obtained for each sample date. These were
then averaged over the sample period to obtain a

2004 Blackwell Publishing Ltd, Freshwater Biology, 50, 105–112
Species diversity and biomass stability 107
single measure of mean taxonomic richness for each
pond.
Previous analyses have shown that measures of
pond productivity, such as total phosphorus, chloro-
phyll a and seston C : N : P strongly covary in these
systems (Steiner, 2004). Temperature, pH and oxygen
also covary, independently of productivity measures
(Steiner, 2004). To reduce collinearities, I used princi-
pal components analysis (PCA), based on the corre-
lation matrix, to create composite variables. Included
in the PCA were time averages of pH, temperature,
oxygen concentration, seston C : P, seston N : P and
log10 transformed measures of total phosphorus, total
chlorophyll a, <35 lm chlorophyll a and >35-lm
chlorophyll a. To explore effects of variable zooplank-
ton composition among ponds, I performed a separate
PCA based on the covariance matrix of log10 (x + 1)
transformed biomass measures of the major zooplank-
ton groups encountered in the survey (a constant was
added because of zero values). I focused on the most
frequently occurring groups, averaging biomass over
the sample period for each pond. These included
Daphnia pulex Leydig, Diaphanosoma brachyurum
Liévin, Ceriodaphnia spp., Scapholeberis spp., Bosmina
logirostris Müller, Chydorus sphaericus Müller, Alona
spp., Pleuroxus spp., calanoid copepods, cyclopoid
copepods, rotifers (all taxa combined), and Chaoborus
spp. To examine the effect of weak interactors (grazer-
resistant algal-prey) on biomass stability, I calculated
mean relative abundance of the >35-lm chlorophyll a
fraction (calculated as mean >35-lm chlorophyll a
divided by mean total chlorophyll a). Relationships
between biomass variability (the CV) and explanatory
variables were explored using ordinary least squares
linear regressions.
Results
A total of 79 zooplankton taxa was encountered in the
survey. As predicted by theory, biomass stability
increased with zooplankton taxonomic richness. Tem-
poral variation in zooplankton biomass, as measured
by the coefficient of variation, was negatively related
to mean zooplankton taxonomic richness (Fig. 1;
R2 ¼ 0.394, P ¼ 0.005).
Principal components analysis of environmental
variables generated two PCA axes accounting for 54.2
and 24.1% of variation in the data set. The first PCA
axis represented a general productivity gradient with
108 C.F. Steiner

2.0 Table 2 Results of the PCA of log10 transformed zooplankton

CV of total zooplankton biomass biomass. Shown are axis loadings.

Variable Axis 1

1.5 Daphnia pulex 0.786*

Diaphanosoma brachyurum )0.698*
Ceriodaphnia )0.083
Chydorus sphaericus )0.130
Scapholeberis 0.007
1.0 Bosmina longirostris )0.817*
Alona 0.001
Pleuroxus 0.003
Calanoid copepods )0.327
0.5 Cyclopoid copepods 0.170
Rotifers 0.016
Chaoborus 0.532*
R 2 = 0.394, P = 0.005
*Variables that were significantly correlated with each axis
0.0
10 15 20 25 (P < 0.05, Bonferroni adjusted).

Mean taxonomic richness

Fig. 1 The relationship between the temporal coefficient of
variation (CV) of total zooplankton biomass and mean
zooplankton taxonomic richness in 18 fishless ponds. The 95%
confidence interval is shown.
total phosphorus, total chlorophyll a, <35 lm chloro-
phyll a and >35-lm chlorophyll a correlating posi-
tively and seston C : P and N : P correlating
negatively with axis scores (Table 1; P < 0.05, Bonfer-
roni adjusted). Oxygen concentration, pH and tem-
perature were negatively correlated with the second
PCA axis (Table 1; P < 0.05, Bonferroni adjusted).
PCA of zooplankton composition produced a single
PCA axis accounting for 51.5% of variation in the
data. Daphnia pulex and Chaoborus were strongly and
positively correlated with axis 1 (Table 2). Diaphano-
soma and Bosmina were strongly and negatively
correlated with axis 1 (Table 2).
Table 1 Results of the PCA of pond environmental variables.
Shown are axis loadings.
Variable
Log10TP
Log10Total Chl-a
Log10<35-lm Chl-a
Log10>35-lm Chl-a
Seston C : P
Seston N : P
pH
Temperature
Oxygen concentration
*Variables that were significantly correlated with each axis
(P < 0.05, Bonferroni adjusted).
I used multiple linear regressions to explore the
effects of mean taxonomic richness, mean relative
abundance of >35-lm chlorophyll a, environmental
PCA axes 1 and 2, and zooplankton PCA axis 1 on the
CV of total zooplankton biomass. When running the
complete regression model, none of the independent
variables were significant at the P < 0.05 level (all
P > 0.10), although the entire model was significant at
P ¼ 0.052 (R2 ¼ 0.560). In both forward and backward
stepwise regressions, only mean zooplankton taxo-
nomic richness was retained in the model (R2 ¼ 0.394,
P ¼ 0.005). The low number of ponds may have
limited the power to detect significant relationships.
As a precaution, I ran separate linear regressions with
each of the explanatory variables. The CV of
zooplankton biomass was unrelated to either
zooplankton PCA axis 1 (R2 ¼ 0.092, P ¼ 0.22) or
environmental PCA axis 2 (pH, temperature and
oxygen) (R2 ¼ 0.001, P ¼ 0.91). The biomass CV
decreased with increasing relative abundance of
>35-lm chlorophyll a (Fig. 2; R2 ¼ 0.317, P ¼ 0.014)
Axis 1 Axis 2 and tended to decrease with increasing values of
0.926* 0.029
environmental PCA axis 1 (pond productivity) (Fig. 3;
0.946* )0.115 R2 ¼ 0.204, P ¼ 0.060).
0.912* )0.214
0.889* 0.055
)0.851* 0.180 Discussion
)0.670* 0.426
)0.457 )0.644* Consistent with model predictions, temporal variation
)0.319 )0.790* in zooplankton biomass decreased with increasing
)0.130 )0.923* taxonomic richness. As shown theoretically, numer-
ous mechanisms may underlie this relationship. First,
Doak et al. (1998) and Tilman et al. (1998) showed that

2004 Blackwell Publishing Ltd, Freshwater Biology, 50, 105–112

2.0
CV of total zooplankton biomass
1.5
1.0
0.5
R 2 = 0.317, P = 0.014
0.0
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8
Mean relative abundance of >35-µm Chl a
Fig. 2 The relationship between the temporal coefficient of
variation (CV) of total zooplankton biomass and mean relative
abundance of chlorophyll a of algae greater than 35 lm in size
(the grazer resistant fraction) for all 18 ponds. The 95% confid-
ence interval is shown.
2.0
CV of total zooplankton biomass
1.5
1.0
0.5
R 2 = 0.204, P = 0.060
0.0
–2 –1 0 1
Environmental PCA 1 (productivity)
Fig. 3 The relationship between the temporal coefficient of
variation (CV) of total zooplankton biomass and environmental
PCA axis 1 (increasing pond productivity) for all 18 ponds. The
95% confidence interval is shown.
species richness effects can occur in the absence of
competition and species interactions because of sta-
tistical averaging or the ‘portfolio effect’. They

2004 Blackwell Publishing Ltd, Freshwater Biology, 50, 105–112
Species diversity and biomass stability 109
1.000
Summed variances
0.100
0.010
0.001
10 15 20 25
Mean taxonomic richness
Fig. 4 The relationship between the sum of temporal variances
in zooplankton biomass (summed across taxa) and mean
zooplankton taxonomic richness. The 95% confidence interval is
shown.
assumed that temporal variance in biomass of each
species i scales with its biomass (mi) as a power
function: variancei ¼ cmzi , where c is a constant and z
is a scaling factor. In the absence of species interac-
tions and assuming that species fluctuations are
independent of one another, the variance of consumer
biomass summed across all species in a community
is a nonlinear function of species richness (N) :
Rvariance ¼ cmzN1)z (for details see Doak et al.,
1998; Tilman et al., 1998; Hughes et al., 2002). Hence,
summed variances will decrease, and stability will
increase, with increasing species richness when z > 1.
Using nonlinear regressions, z in my ponds was
estimated to be 1.447 (R2 ¼ 0.969, P < 0.05) and
summed variances were negatively related to taxo-
nomic richness on the log-log scale (Fig. 4; r ¼ )0.484,
P ¼ 0.042), indicating that portfolio effects were
occurring. This z estimate is similar to previous
2
studies that have shown that z commonly ranges
between 1 and 2 (Tilman, 1999).
Strong effects of interspecific competition are well-
documented among zooplankton species (reviewed in
DeMott, 1989) and are known to occur in a subset of
the ponds included in the present investigation
(Steiner, 2003b). When analysing temporal covari-
ances of zooplankton species biomass in the study
ponds, negative terms were frequent and summed
covariances averaged across ponds tended to be
110 C.F. Steiner
negative (mean ¼ )0.0024); although covariances
averaged across ponds did not differ from zero
(P ¼ 0.872, one-way t-test). Negative covariances are
indicative of competition effects. Alternatively, neg-
ative covariances could be driven by species respond-
ing differently to external perturbations (Yachi &
Loreau, 1999). Regardless of underlying mechanisms,
negative covariances will increase stability of zoo-
plankton communities by reducing net variances in
total community biomass. However, in order for such
covariances to account for the observed stability-
richness relationship (Fig. 1), the sum of covariances
must become more negative with increasing species
richness. This was not the case in the present data set;
no relationship between mean taxonomic richness and
summed covariances was apparent among ponds
(r ¼ 0.039, P ¼ 0.877). Thus, portfolio effects are a
more likely explanation for observed patterns.
One potential criticism of this study is that several
zooplankton taxa were not identified to the species
level. The measures of taxonomic richness were meant
as correlates or indices of true species richness and
would fail only if certain taxa were highly speciose.
As the majority of zooplankton was identified to a
high level of taxonomic resolution (either species or
genus), the probability of a strong bias in the data set
seems minimal. Of greatest concern are the copepods,
in which course levels of identification were used
(either calanoid or cyclopoid). However, previous
studies that have identified copepods to the level of
species in these ponds have consistently found only
one species of calanoid copepod and up two species of
cyclopoid copepod (J. Shurin, University of British
Columbia, personal communication). Hence, it seems
unlikely that a finer taxonomic resolution would
greatly alter the results. Another potential criticism
is that microzooplankton, such as protozoa and small
species of rotifers that passed through the sampling
sieve, were not included in measures of taxonomic
richness. Inclusion of such taxa would affect the
results if species richness within these groups varied
systematically among the ponds. As I am unable to
assess the magnitude of this potential problem, some
caution is warranted. Finally, zooplankton sample
volumes varied among ponds. Thus, it is possible that
rare zooplankton taxa were missed in some samples,
potentially confounding species richness effects with a
sample size effect. However, in the year following the
survey, collector curves comparing the total number
of zooplankton taxa encountered with total sample
volume were generated for a number of the study
ponds spanning a range of productivities. These
revealed that sample volumes of 10–24 L were often
adequate for detecting the majority (>80%) of
zooplankton taxa. Sample volumes collected for the
present investigation were all equal to or greater than
24 L. Finally, no relationship between mean zooplank-
ton taxonomic richness and mean sample volume was
detected either on the arithmetic scale (P ¼ 0.33,
Pearson correlation) or logarithmic scale (P ¼ 0.12,
Pearson correlation), further indicating that a samp-
ling artefact was unlikely.
Although biomass stability increased with
zooplankton taxonomic richness, numerous factors
covaried among the study ponds and may have
influenced observed variation. Thus, the diversity
effect is not the only viable explanatory hypothesis.
Regressions showed that temporal stability of biomass
increased with increasing relative abundance of
grazer-resistant algae (>35-lm chlorophyll a). This
observation is qualitatively consistent with recent
investigations that have shown that the presence of
weak interactors and inedible prey can stabilise
community dynamics (Abrams & Walters, 1996;
McCann et al., 1998; Bohannan & Lenski, 1999;
McCauley et al., 1999). However, enrichment is also
thought exert an important destabilising influence on
community dynamics, causing stable consumer-
resource dynamics to become temporally variable
(Rosenzweig, 1971; Gilpin, 1972). My results were
consistent with this hypothesis as well; the CV of
zooplankton biomass increased (i.e. stability de-
creased) with increasing system enrichment.
A handful of empirical studies has examined the
relationship between species diversity and ecosystem
stability. Although there are exceptions, many of these
investigations show that the relationship between
temporal variability of community-level biomass and
species richness is negative (reviewed in Loreau et al.,
2002; Schmid, Joshi & Schlapfer, 2002). This study
provides further evidence that diversity may play an
important role in maintaining the temporal stability of
aggregate community measures. However, this con-
clusion must be tempered by the caveat that other
environmental variables in the study ponds may have
driven the diversity-stability relationship. Making
inferences from natural pattern can be problematic
as numerous factors commonly covary and exhibit

2004 Blackwell Publishing Ltd, Freshwater Biology, 50, 105–112

collinearity among sample sites. In my study system,
it is difficult to assess the relative importance of
potential driving mechanisms. While one of the
above-named factors (taxonomic richness, productiv-
ity, or the prevalence of grazer resistant algae) may
have been the true mediator of biomass stability, it is
also possible that all three interacted to generate
observed variability. Only direct experimental manip-
ulations in conjunction with proper control of extra-
neous environmental factors will offer stronger
insights.
Acknowledgments
Gary Mittelbach, Alan Tessier, Steve Hamilton, Jim
Bence and Don Hall provided valuable advice
during the planning and execution of this research.
I also thank Alan Hildrew, James Grover and an
anonymous reviewer for comments on the manu-
script. I was supported by a Sigma Xi Grant-in-Aid
of Research, the K.B.S. Graduate Research Training
Group (RTG) funded by National Science Foundation
grants DIR-09113598 and DBI-9602252, and a N.S.F.
Microbial Biology Postdoctoral Fellowship during
the writing of the paper. This is contribution number
1160 of the W. K. Kellogg Biological Station,
Michigan State University.
References
Abrams P.A. & Walters C.J. (1996) Invulnerable prey and
the paradox of enrichment. Ecology, 77, 1125–1133.
Bohannan B.J.M. & Lenski R.E. (1999) Effect of prey
heterogeneity on the response of a model food chain to
resource-enrichment. American Naturalist, 153, 73–82.
Carpenter S.R., Kitchell J.F., Hodgson J.R., Cochran P.A.,
Elser J.J., Elser M.M., Lodge D.M., Kretchmer D. & He
X. (1987) Regulation of lake primary productivity by
food web structure. Ecology, 68, 1863–1876.
Cottingham K.L. (1999) Nutrients and zooplankton as
multiple stressors of phytoplankton communities:
evidence from size structure. Limnology and Oceanog-
raphy, 44, 810–827.
Cottingham K.L., Brown B.L. & Lennon J.T. (2001)
Biodiversity may regulate the temporal variability of
ecological systems. Ecology Letters, 4, 72–85.
DeMott W.R. (1989) The role of competition in zooplank-
ton succession. In: Plankton Ecology: Succession in
Plankton Communities (Ed. U. Sommer), pp. 195–252.
Springer-Verlag, New York.

2004 Blackwell Publishing Ltd, Freshwater Biology, 50, 105–112
Species diversity and biomass stability 111
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 com-
munity ecology. American Naturalist, 151, 264–276.
Gilpin M.E. (1972) Enriched predator-prey systems:
theoretical stability. Science, 177, 902–904.
Hughes J.B., Ives A.R. & Norberg J. (2002) Do species
interactions buffer environmental variation (in
theory)? In: Biodiversity and Ecosystem Functioning:
Synthesis and Perspectives (Eds M. Loreau, S. Naeem &
P. Inchausti), pp. 92–101. Oxford University Press,
New York.
Ives A.R. & Hughes J.B. (2002) General relationships
between species diversity and stability in competitive
systems. American Naturalist, 159, 388–395.
Ives A.R., Gross K. & Klug J.L. (1999) Stability and
variability in competitive communities. Science, 286,
542–544.
Ives A.R., Klug J.L. & Gross K. (2000) Stability and
species richness in complex communities. Ecology
Letters, 3, 399–411.
Loreau M., Downing A., Emmerson M., Gonzalez A.,
Hughes J., Inchausti P., Joshi J., Norberg J. & Sala O.
(2002) A new look a the relationship between diversity
and stability. In: Biodiversity and Ecosystem Functioning:
Synthesis and Perspectives (Eds M. Loreau, S. Naeem &
P. Inchausti), pp. 79–91. Oxford University Press, New
York.
McCann K.S. (2000) The diversity-stability debate.
Nature, 405, 228–233.
McCann K., Hastings A. & Huxel G.R. (1998) Weak
trophic interactions and the balance of nature. Nature,
395, 794–798.
McCauley E. (1984) The estimation of the abundances
and biomass of zooplankton in samples. In: A Manual
on the Methods for the Assessment of Secondary Produc-
tivity in Freshwaters (Eds J.A. Downing & F. Rigler), pp.
228–265. Blackwell Scientific Publications, Oxford.
McCauley E. & Murdoch W.W. (1990) Predator-prey
dynamics in environments rich and poor in nutrients.
Nature, 343, 455–457.
McCauley E., Nisbet R.M., Murdoch W.W., de Roos A.M.
& Gurney W.S.C. (1999) Large-amplitude cycles of
Daphnia and its algal prey in enriched environments.
Nature, 402, 653–656.
Petchey O.L. (2000) Prey diversity, prey composition,
and predator population dynamics in experimental
microcosms. Journal of Animal Ecology, 69, 874–882.
Rosenzweig M.L. (1971) Paradox of enrichment: desta-
bilization of exploitation ecosystems in ecological time.
Science, 171, 385–387.
Sarnelle O. (1992) Nutrient enrichment and grazer effects
on phytoplankton in lakes. Ecology, 73, 551–560.
112 C.F. Steiner
Schmid B., Joshi J. & Schlapfer F. (2002) Empirical
evidence for biodiversity-ecosystem functioning rela-
tionships. In: The Functional Consequences of Biodiversity:
Empirical Progress and Theoretical Extensions (Eds A.P.
Kinzig, S.W. Pacala & D. Tilman), pp. 120–150.
Princeton University Press, Princeton.
Steiner C.F. (2002) Context-dependent effects of Daphnia
pulex on pond ecosystem function: observational and
experimental evidence. Oecologia, 131, 549–558.
Steiner C.F. (2003a) Keystone predator effects and grazer
control of planktonic primary production. Oikos, 101,
569–577.
Steiner C.F. (2003b) Variable dominance in pond com-
munities: assessing spatiotemporal variation in com-
petition and predation intensity. Ecology, 84, 982–990.
Steiner C.F. (2004) Daphnia dominance and zooplankton
community structure in fishless ponds. Journal of
Plankton Research, 26, 799–810.
Sterner R.W. (1989) The role of grazers in phytoplankton
succession. In: Plankton Ecology: Succession in Plankton
Communities (Ed. U. Sommer), pp. 107–169. Springer-
Verlag, New York.
Tilman D. (1999) The ecological consequences of changes
in biodiversity: a search for general principles. Ecology,
80, 1455–1474.
Tilman D., Lehman C. & Bristow C.E. (1998) Diversity-
stability relationships: statistical inevitability or ecolo-
gical consequence? American Naturalist, 151, 277–282.
Yachi S. & Loreau M. (1999) Biodiversity and ecosystem
productivity in a fluctuating environment: the insur-
ance hypothesis. Proceedings of the National Academy of
Sciences, 96, 1463–1468.
(Manuscript accepted 30 September 2004)

2004 Blackwell Publishing Ltd, Freshwater Biology, 50, 105–112