Download COMPETITOR-INDUCED PLASTICITY IN TADPOLES

Ecological Monographs, 72(4), 2002, pp. 523–540
q 2002 by the Ecological Society of America
COMPETITOR-INDUCED PLASTICITY IN TADPOLES: CONSEQUENCES,
CUES, AND CONNECTIONS TO PREDATOR-INDUCED PLASTICITY
RICK A. RELYEA1
Department of Biological Sciences, University of Pittsburgh, Pittsburgh, Pennsylvania 15260 USA
Abstract. Phenotypically plastic responses to environmental change are typically compartmentalized by the type of environmental cues that cause the induction (e.g., temperature,
light), but different types of environmentally induced responses might very well be related
to each other. I used a number of experiments to document competitor-induced behavior
and morphology in wood frog tadpoles ( Rana sylvatica), to quantify how competitor-induced
changes affected subsequent competitive ability, and to investigate the cues responsible for
the competitor-induced response. Competitors induced wood frogs to increase their activity
and generally develop larger bodies and smaller tails. Further, wood frogs were able to
discriminate between intraspecific and interspecific competitors. Whereas behavioral responses to competitors are well documented, the widespread morphological responses are
a new discovery; these responses are particularly intriguing because they are in the opposite
direction of predator-induced responses in tadpoles. The competitor-induced phenotypes
are not simply an allometric effect, but appear to be adaptive responses to competitors;
competitor-induced wood frogs experienced higher relative growth rates than noninduced
wood frogs in subsequent performance trials, but past studies suggest that the competitorinduced phenotype should experience higher predation risk. When responding to competition, wood frogs were not responding only to reduced food. They were able to sense
changes in both the per capita food levels and changes in density (per unit volume) that
were independent of per capita food level. Thus, these animals have an amazing ability to
sense changes in their environments and respond in very precise ways.
When one considers competitor-induced responses in light of past studies on predatorinduced phenotypes in larval anurans, one sees that predator-induced traits provide individuals with increased predator resistance but decreased competitive ability, whereas competitor-induced traits provide individuals with increased competitive ability but decreased
predator resistance. This suggests that predator- and competitor-induced plasticity have
evolved an intricate link that presents a trade-off between competitive ability and predator
resistance ability. This trade-off is common across species and makes sense in light of the
environments in which wood frogs live, ranging from ponds with high densities of predators
and few tadpoles competing to ponds with few predators and extremely high densities of
tadpoles competing. Further, there is evidence that this type of plasticity trade-off might
be quite common in other phenotypically plastic taxa.
Key words: adaptive plasticity; anura; behavior; competition; cues; morphology; phenotypic
plasticity; Pseudacris triseriata; Rana pipiens; Rana sylvatica; tadpole.
INTRODUCTION
Phenotypic plasticity allows organisms to respond to
environmental change by altering their behavior, morphology, physiology, or life history. Over the years,
ecologists have amassed a large literature documenting
the existence of phenotypically plastic traits in a wide
variety of taxa responding to a host of biotic and abiotic
environments (Bradshaw 1965, Schlichting 1986, Sultan 1987, West-Eberhard 1989, Schlichting and Pigliucci 1998). While plastic responses are not necessarily adaptive, there is increasing evidence that many
environmentally induced responses are at least in adaptive directions such that the phenotype expressed in a
given environment experiences higher fitness than any
Manuscript received 3 November 2000; revised 11 June 2001;
accepted 19 June 2001; final version received 9 November 2001.
1 E-mail: relyea 1 @pitt.edu
alternative phenotype (the ‘‘adaptive plasticity hypothesis’’; Dudley and Schmitt 1996). While we continue
to add to the list of organisms and the kinds of traits
that are phenotypically plastic, a current challenge is
to begin examining how different types of phenotypically plastic responses are related to each other and
how these interrelated responses may have evolved.
One type of phenotypic plasticity that is ubiquitous
in nature is predator-induced plasticity. Many plants
respond to herbivory by facultatively producing antiherbivore chemicals (Karban and Baldwin 1997), and
many animals respond to predators by altering their
behavior (e.g., through spatial avoidance or activity
reduction; Lima and Dill 1990, Kats and Dill 1998) or
by altering their morphology (e.g., producing attackdeterring structures or escape structures; Havel 1987,
Harvell 1990, Tollrian and Harvell 1999). These predator-induced traits typically result in decreased pre-
523
524
RICK A. RELYEA
dation risk but often come at the cost of decreased
growth and development when predators are absent.
A second ubiquitous type of phenotypic plasticity is
competitor-induced plasticity. In plants, increased
competition for resources commonly induces morphological changes. For example, plants typically respond
to competition for light by decreasing their root:shoot
ratio and respond to competition for soil nutrients by
increasing their root:shoot ratio (Barrett and Wilson
1981, Schlichting and Levin 1984, Dudley and Schmitt
1996). Animals often respond to competition by increasing their foraging activity and altering their diet
(Akre and Johnson 1979, Crowley 1979, Formanowicz
1982, Inoue and Matsura 1983, Stephens and Krebs
1986, Anholt and Davies 1987, Macchiusi and Baker
1992, Horat and Semlitsch 1994, Anholt and Werner
1995, Anholt et al. 1996). We have a few examples of
animals altering their morphology in response to competition. Several marine invertebrates (anemones, corals, octocorals, hydroids, bryozoans, and rotifers) produce defensive structures that are effective in deterring
interference competition (Stemberger and Gilbert
1987, Harvell 1990). In addition, there are a few amphibians that inhabit highly ephemeral habitats and
change from an omnivorous to a carnivorous (and often
cannibalistic) phenotype under high intraspecific competition (these cases appear to be limited to only a few
species of amphibians; Bragg 1956, Reilly et al. 1992,
Pfennig 1992a,b, Pfennig and Collins 1993). In general, competitor-induced responses result in more rapid
growth and improved competitive ability but, at least
in the case of behavior, come at the cost of increased
predation risk (e.g., Anholt and Werner 1995).
When we consider predator- and competitor-induced
responses simultaneously, it becomes apparent that the
two types of plasticity might be intricately linked and
traded off. Such a trade-off has been repeatedly observed between species (Paine 1966, Morin 1981, Werner and McPeek 1994) to the point that it has become
a paradigm in community ecology (Vance 1978,
Abrams 1982, Werner and Anholt 1993, Holt and Lawton 1994, Leibold 1996). A link between predator- and
competitor-induced plasticity can be found in the activity level exhibited by animals; predators induce lower activity, which makes prey more resistant to predation but less competitive whereas competitors induce
higher activity, which makes prey more competitive
but less resistant to predation (Werner and Anholt 1993,
McNamara and Houston 1994). Similarly, morphological responses to predators make prey more resistant
to predation at the cost of slower growth and development (which likely translates to reduced competitive
ability). As noted above, we have few examples of
morphological responses to competitors. If morphological responses follow the pattern of behavioral responses, we would predict that competitors induce morphological changes that make prey more competitive
but less resistant to predation. Thus, competitors and
Ecological Monographs
Vol. 72, No. 4
predators should induce the same suite of behavioral
and morphological traits, but in opposite directions. To
test this hypothesis, one should search for competitorinduced morphological plasticity in taxa that are known
to exhibit predator-induced morphological plasticity.
Documenting competitor-induced behavioral and
morphological plasticity is important in addressing the
hypothesis, but it is critical that we also determine
whether the competitor-induced plasticity is adaptive.
To be adaptive, the competitor-induced phenotype
should experience superior fitness (e.g., more rapid
growth) in a competitive environment relative to other
phenotypes and inferior fitness in alternative environments (Via and Lande 1985, 1987, Via 1987, Van Tienderen 1991, Gomulkiewicz and Kirkpatrick 1992).
Support of this hypothesis does not necessarily indicate
that natural selection caused the current level of plasticity, but it does support the hypothesis that natural
selection is currently maintaining the existing plasticity
(Van Buskirk and Relyea 1998).
For adaptive plasticity to evolve, there also must be
reliable cues that indicate the current state of the environment (Dodson 1989). For competitor-induced responses, the most obvious cue is the per capita resource
level (e.g., the concentration of food). Declining per
capita resources should serve as an index to the intensity of competition for resources both currently and in
the immediate future. There are several theoretical and
empirical behavioral studies that support this mechanism (Macchiusi and Baker 1992, Werner and Anholt
1993, McNamara and Houston 1994, Anholt and Werner 1995, Anholt et al. 1996). Alternatively (or additionally), individuals might respond to increased densities of competitors independently of per capita resource levels. I am aware of no studies that have distinguished between these two possibilities.
I tested these hypotheses using a system of larval,
pond-dwelling anurans. Larval anurans in the midwestern United States enter a pond as oviposited eggs
and then quickly hatch into larvae (tadpoles). After a
period of growth (spanning weeks to years), tadpoles
metamorphose into terrestrial frogs and leave the pond.
Larval anurans have a well-documented ability to exhibit predator-induced plasticity; in the presence of
aquatic predators (e.g., aeshnid dragonflies and ambystomatid salamanders), many species become less
active and develop relatively deeper tails and smaller
bodies compared to conspecifics reared without predators. Predator-induced tadpoles survive predation better but they grow more slowly when predators are absent (Smith and Van Buskirk 1995, McCollum and Van
Buskirk 1996, McCollum and Leimberger 1997, Van
Buskirk and Relyea 1998, Relyea 2001c, 2002a). With
competition, tadpoles become more active but we know
little about what happens to their morphology (but see
Relyea 2001b). My goal was to examine how competition alters both behavior and morphology, to test
how competitor-induced changes affect the competitive
COMPETITOR-INDUCED TADPOLES
November 2002
performance of the tadpoles, and to identify the cues
responsible for the changes. More specifically, my hypotheses were the following: (1) competitors induce
behavioral and morphological changes in the opposite
direction from previously observed predator-induced
changes, (2) intraspecific and interspecific competition
cause similar phenotypic responses, (3) competitor-induced responses make tadpoles more competitive (i.e.,
the responses are adaptive), and (4) the cues that induce
the responses are reductions in per capita food levels.
I conducted three separate experiments to test the
four hypotheses. The first experiment examined how
wood frogs alter their behavioral and morphological
phenotypes in the presence of intraspecific and interspecific competition. The second experiment examined
how competitor- and predator-induced changes affected
subsequent competitive ability of wood frog tadpoles.
The third experiment examined whether the inducing
cues are changes in food level, changes in tadpole density, or both.
METHODS
Experiment 1: Competitor-induced
phenotypic changes
To determine how wood frogs altered their traits with
increased intraspecific and interspecific competition, I
raised wood frog tadpoles (Rana sylvatica) with different numbers of intra- and interspecific competitors.
This experiment was conducted at the Experimental
Pond Facility on the University of Michigan’s Edwin
S. George Reserve (ESGR) in southeast Michigan. I
used a randomized block design (two spatial blocks,
one replicate per block) with a factorial combination
of three treatments within each block. The three treatments were competitor species (intraspecific or interspecific), total competitor density (20, 40, or 60 tadpoles), and wood frog population (Buffer Zone Marsh,
Cattail Marsh, Southwest Woods Pond, and Dreadful
Hollow; all located on the ESGR), for a total of 24
treatments. I used multiple wood frog populations because previous work demonstrated that wood frogs can
have population-specific responses to environmental
change (Relyea 2002b). Intraspecific competition was
manipulated using increasing intrapopulation densities
of wood frog larvae (20 wood frogs 5 25 larvae/m2,
40 wood frogs 5 50 larvae/m2, and 60 wood frogs 5
75 larvae/m2). Natural densities can reach up to 400
wood frog tadpoles/m2 (E. E. Werner, R. A. Relyea, D.
K. Skelly, and K. L. Yurewicz, unpublished data). Interspecific competition was manipulated using increasing densities of chorus frog larvae (Pseudacris triseriata; 20 wood frogs 1 0 chorus frogs, 20 wood frogs
1 20 chorus frogs, and 20 wood frogs 1 40 chorus
frogs). Chorus frog larvae naturally coexist with wood
frog larvae (Collins and Wilbur 1979).
All animals were initially collected as eggs from
natural ponds on the ESGR. Wood frogs were collected
525
from Buffer Zone Marsh, Cattail Marsh, Southwest
Woods Pond, and Dreadful Hollow on 4 April 1997,
and chorus frogs were collected from Buffer Zone
Marsh on 10 April and 13 April 1997. Eggs were
hatched in wading pools containing well water and the
tadpoles were fed rabbit food ad libitum until used in
the experiments. The four wood frog populations began
the experiment at similar initial mean masses (Buffer
Zone Marsh, 16 mg; Cattail Marsh, 12 mg; Southwest
Woods Pond, 15 mg; Dreadful Hollow Pond, 10 mg)
and chorus frogs began the experiment at 19 mg.
The experimental mesocosms were 100-L plastic
wading pools filled with well water, 100 g of leaves
(primarily Quercus spp.), 5 g of rabbit chow, and an
aliquot of zooplankton and phytoplankton from a nearby pond. All pools were covered with a continuous
piece of 60% shade cloth suspended over the pools to
create a screened ‘‘tunnel’’ that excluded predators and
ovipositing amphibians. I began the experiment on 8
May 1997 and continued it until the chorus frogs approached metamorphosis (Gosner stage 42; Gosner
1960) on 17 June 1997 (40 d from the start of the
experiment). During the experiment, I observed the activity of the tadpoles 20 times during all hours of the
day by making scan samples of the tadpoles on the
sides and bottom of each pool (Altmann 1974). During
each scan, I counted the number of wood frog tadpoles
that could be observed and the number of observed
tadpoles that were moving. Dividing the number of
moving tadpoles by the number of observed tadpoles
provided an activity percentage. I averaged the 20 observations from each pool to provide a single estimate
of activity percentage for each pool. Upon termination
of the experiment, the tadpoles were removed from the
pools, separated by species, counted, and weighed. The
wood frog larvae were then preserved in 10% formalin
for subsequent morphological analysis. Chorus frogs
were not preserved because they were needed for other
experiments.
I measured the morphology of the wood frog tadpoles using computer-aided image analysis software
(Bioscan Optimas, Bothell, Washington), which allowed me to project video images of preserved tadpoles
onto a monitor and make seven linear measurements.
From the lateral view, I measured the tail length, maximum tail depth, maximum tail muscle depth, maximum body length, and maximum body depth; from the
dorsal view, I measured the maximum tail muscle
width, the maximum body width, and the maximal
mouth width (the widest denticle row; for a photograph
of dimensions, see Relyea 2000). To make tadpoles lay
in a more natural position, I placed a piece of glass
under the tail so that it was parallel with the midline
of the body. For each pool, 10 preserved tadpoles were
measured.
Statistical analysis.—The effect of competition on
wood frog tadpoles was analyzed using two multivariate analyses of variance (MANOVAs). The first MAN-
RICK A. RELYEA
526
OVA examined the effects of block, population, competitor density, and competitor species (wood frog or
chorus frog) on wood frog survivorship, daily growth
rate ([final mass 2 initial mass]/d), activity, and relative morphology. The second MANOVA was similar
to the first, except that it examined the effect of competitor intensity rather than competitor density, allowing me to investigate how behavior and morphology
were affected on a per-unit-of-competition basis. I
quantified competitive intensity as
CI 5 (G0 2 GC)/G0
where CI is competitive intensity, G0 is the growth rate
of wood frogs with zero additional competitors, and
GC is the growth rate of wood frogs at a given density
of competitors. Thus, competitive intensity can range
from zero (no loss of growth relative to the low density
treatment) to one (a total loss of growth under extremely high competition).
To examine relative differences in morphology, I had
to first remove differences in morphology that were
due to differences in overall size. To this end, I regressed the seven linear measures (log-transformed to
improve the linearity of the relationships) against the
log-transformed mass of each individual. The residuals
of this regression provided estimates of relative differences in morphology (Bookstein 1989). I then calculated the mean residual from each pool and used
these mean residuals as the responses in the MANOVA.
For all analyses, the three-way, four-way, and blockby-treatment interactions were nonsignificant and
pooled with the error term to increase power of the
tests. For chorus frogs, only survivorship was analyzed
using a two-way ANOVA.
Experiment 2: Competitive ability of competitor- and
predator-induced tadpoles
The second experiment examined the growth performance of both competitor- and predator-induced
wood frogs using a two-stage procedure that first created alternative phenotypes and then assessed the relative performance of each phenotype. This experiment
also was conducted at the ESGR. In the first stage, 20
cattle watering tanks were filled with 1000 L of well
water on 28 April 1998. I then added 300 g of dry
leaves (primarily Quercus spp.), 25 g of rabbit chow,
and a 1-L aliquot of pond water to serve as a source
of phytoplankton and periphyton (Werner and Anholt
1996). The following day, I added a concentrated aliquot of zooplankton to each tank. Thus, similar to experiment 1, each tank contained many components of
a natural pond.
The experimental design was a completely randomized design that used a factorial combination of low
and high competition crossed with the absence and
caged presence of predatory dragonfly larvae (Anax
longipes). The four treatments were replicated five
times. On 11 May 1998, 75 newly hatched wood frog
Ecological Monographs
Vol. 72, No. 4
tadpoles from the nearby Doyle Road population (mean
mass 6 1 SE 5 69 6 4 mg) were stocked per tank (37
m2). The low and high competition treatments involved
the addition of zero and 75 leopard frog (Rana pipiens)
tadpoles, respectively (mean mass 6 1 SE 5 71 6 5
mg). Wood frogs were a mixture of $20 egg masses
whereas leopard frogs were a mixture of 4 egg masses.
A sample was set aside to assess handling survival;
both species experienced 100% survival after 24 h.
All tanks were equipped with four predator cages,
constructed of 10 3 10 cm drain pipe with nylon windowscreen on either end. A single penultimate-stage
Anax was placed into each cage for tanks assigned the
predator treatment; no-predator tanks contained empty
cages. Each predator was fed ;300 mg of wood frogs
(3–4 tadpoles) three times per week. Empty cages were
lifted and replaced to equalize disturbance across all
treatments. All tanks were covered with 60% shadecloth lids to prevent colonization by insect predators
and other ovipositing anurans. On 3 June, 30 wood frog
tadpoles were removed from each of the 20 tanks; 10
of the tadpoles were preserved in 10% formalin for
subsequent morphological analysis and 20 of the tadpoles were set aside for the second stage of the experiment.
The second stage of the experiment employed 100L pond mesocosms to determine the relative performance of the four groups of wood frogs induced in the
first stage. On 1 June, I filled 40 100-L wading pools
with well water and then added 100 g oak leaves, 5 g
rabbit chow, and aliquots of plankton as described in
experiment 1. The pools were randomly assigned one
of the four wood frog groups generated in the first stage
of the experiment crossed with the presence or absence
of 20 new leopard frogs (mass 5 227 6 3 mg) for a
total of eight treatments replicated five times. During
the second-stage experiment, I observed the activity of
the tadpoles (as described in experiment 1) 18 times
during all hours of the day from 4–12 June. The average
of the 18 observations was used as the response variable for each pool. Because wood frogs and leopard
frogs are difficult to discriminate while swimming in
pools, the activity observations were restricted to those
pools containing wood frogs reared alone. The experiment was terminated on 12 June; all tadpoles were
sorted from the leaves, counted, and weighed to determine survivorship and relative growth. Relative
growth was calculated as the final mass of the tadpoles
divided by the mass of the tadpoles at the beginning
of the performance trial.
Statistical analysis.—I analyzed the data using
MANOVAs and Wilks’ lambda test criterion (SYSTAT
1992). In the first stage of the experiment, I analyzed
the impact of the competitor and caged-predator environments on the growth rate, activity, and relative
morphology of the sampled wood frog larvae (using
the same size-adjusting protocol for morphology as described in experiment 1). In the second stage of the
November 2002
COMPETITOR-INDUCED TADPOLES
experiment, I analyzed the impacts of phenotype and
subsequent competitor environment on wood frog activity and growth rate. Survivorship was not included
in the analysis because survivorship was very high
across all treatments (mean 5 97% 6 1%).
Experiment 3: Identification of competitor cues
The cues used by tadpoles for exhibiting the suite
of phenotypic responses is unknown but could include
changes in per capita food resources or changes in density. To discriminate between these two possibilities, I
conducted a third experiment at the University of Pittsburgh’s Pymatuning Laboratory of Ecology in northwestern Pennsylvania using a population of wood frogs
collected as eggs from Tryon Woods. I began by rearing
wood frog tadpoles (initial mean mass 5 99 mg) in the
laboratory in 12-L containers filled with aged tap water.
The treatments were either a constant food ration
(ground fish food at a ration of 12% of the total mass
of eight tadpoles) with increasing intraspecific densities
(8, 16, or 32 tadpoles), or a constant density (8 tadpoles) with decreasing food rations (12%, 6%, or 3%).
This design is similar to that used by Wilbur (1977) in
his investigations of growth and developmental plasticity in wood frog larvae. All treatments were replicated four times using four spatial blocks. In this way,
per capita resource levels were at 12%, 6%, and 3%
using either increased density or decreased food. If
tadpoles produce the same phenotypes under both sets
of treatments, then it supports the hypothesis that per
capita food level is the inducing cue. However, if the
tadpoles grow equally under the two treatments but
exhibit different behavior or morphology, then it suggests reduced food and increased density both provide
cues that induce competitive responses. Ground fish
food was selected as a food source because it is very
homogeneous, which eliminates concerns about the
tadpoles under different competitive treatments foraging on heterogeneous foods of varying qualities.
The tadpoles were added to the tubs on 9 May 2000
and allowed to grow for 14 d. Tadpoles were observed
for activity 10 times each day for the first 13 d and the
mean percentage of active tadpoles for the 130 observations was used to represent each tub’s response variable. Upon termination of the experiment, all tadpoles
were weighed and eight tadpoles from each tub were
preserved in 10% formalin for subsequent morphological analysis. All morphological dimensions were measured and made size-independent as described in experiment 1. I analyzed the experiment using a MANOVA that examined the impact of per capita food ration
and method of food ration on the growth rate, activity,
and relative morphology of the wood frog tadpoles.
Survival was very high across all treatments (94–
100%), so I omitted survival from the analysis. The
block effects and block-by-treatment interactions were
nonsignificant and pooled with the error term.
Because the experiment indicated that the tadpoles
527
were using both per capita food ration cues and competitor density cues, I conducted a follow-up experiment to better understand how tadpoles could possibly
detect changes in competitor density independent of
changes in per capita food rations. Tadpoles that detect
an increase in competitors could simply be sensing an
increase in absolute number of tadpoles or an increase
in the relative density of tadpoles (per unit area or per
unit volume). To differentiate between these possibilities, I reared wood frog tadpoles in identical containers
under a factorial combination of two water volumes (6
L and 12 L) and two competitor densities (8 or 16
tadpoles) at a constant food ration (ground Tetramin
[Tetra, Incorporated, Blacksburg, Virginia] at a ration
of 12% of the total body mass of eight tadpoles). If
wood frogs exhibit different activity levels between the
two water volumes at a given density, then it suggests
that tadpoles detect changing competitor densities by
sensing changes in relative density per unit volume.
For this experiment, I used the same population of tadpoles (initial mean mass 5 223 mg), replicated the
experiment four times, and used activity level as a rapid
assay. I observed tadpole activity 10 times per day for
6 d and used the mean of the 60 observations as each
tub’s response variable. I analyzed the data using a twoway analysis of variance.
RESULTS
Experiment 1: Competitor-induced
phenotypic changes
Competition caused widespread effects on the traits
of larval wood frogs. There was a significant multivariate effect of block, density, and a density-by-species interaction, but no effect of wood frog population
(Table 1, Fig. 1). Univariate analyses revealed that increased competition did not affect wood frog survivorship but did reduce wood frog growth. Both intraspecific and interspecific competition reduced growth,
but the intraspecific effect was larger. Intraspecific
competition increased wood frog activity (F1,18 5
36.66, P 5 0.00001), but interspecific competition had
no effect (F1,18 5 0.005, P 5 0.944). Wood frog morphology was affected similarly by the presence of intraspecific and interspecific competitors. Increased
competition caused tadpoles to develop relatively short
and shallow tails, narrow and shallow muscles, long
and deep bodies, and wide mouths. Relative body width
was unaffected.
In the second analysis, I examined the effect of competitive intensity on wood frogs (Table 2, Fig. 2). There
was a significant multivariate effect of block, competitive intensity, and a competitive intensity-by-species
interaction. Similar to the effect of increased competitor density, increased competitive intensity resulted in
increased wood frog activity when conspecifics were
added (F3,15 5 29.98, P 5 0.00006) but not when chorus
frogs were added (F3,15 5 0.01, P 5 0.916). Increased
Ecological Monographs
Vol. 72, No. 4
RICK A. RELYEA
528
TABLE 1. MANOVA results examining how wood frog survivorship, growth rate, activity, and relative morphology
were affected by experimental block, wood frog population,
competitor density, and competitor species (intraspecific or
interspecific; experiment 1).
A) Multivariate tests
Factor
df
F
P
Block
Population
Species
Density
Population 3 Species
Population 3 Density
Species 3 Density
11,24
33,71
11,24
11,24
33,71
33,71
11,24
4.96
1.18
0.93
39.12
1.37
0.89
4.92
0.0005
0.271
0.531
,0.00001
0.135
0.632
0.0005
B) Univariate tests
Factor
Survivorship
Growth rate
Activity
Tail length
Tail depth
Muscle depth
Muscle width
Body length
Body depth
Body width
Mouth width
Density
0.914
,0.00001
0.004
0.0001
0.020
,0.00001
,0.00001
0.002
0.0003
0.087
0.014
Species 3
Density
0.746
0.0004
0.003
0.214
0.272
0.084
0.487
0.861
0.793
0.135
0.748
Notes: All three-way, four-way, and block interactions were
nonsignificant and pooled with the error term. (A) reports the
multivariate tests and (B) reports the univariate tests for all
significant multivariate tests.
competitive intensity also caused smaller and less muscular tails, larger bodies, and wider mouths. However,
in this analysis, there were differences between the
effects of intra- and interspecific intensity on morphological traits. Increased competitive intensity of chorus
frogs had a larger impact on wood frog tail length, tail
depth, muscle width, and body depth than the increased
competitive intensity of wood frogs. Chorus frog survivorship did not differ among treatments (P $ 0.1).
Experiment 2: Competitive ability of competitor- and
predator-induced tadpoles
Competitors and predators, as well as their interaction, had significant multivariate effects on wood frog
phenotypes (Table 3, Fig. 3). Competitors significantly
reduced growth while caged predators had no effect.
In the absence of predators, interspecific competition
induced deeper tails as well as long, shallow, and narrow bodies. In the presence of predators, competition
continued to induce long, shallow, and narrow bodies
but it also induced longer and shallower tails, wider
tail muscles, and higher activity. In the absence of competitors, predators induced shorter and deeper tails,
deeper tail muscles, shorter bodies, and lower activity.
In the presence of competitors, predators induced deeper tails, deeper and wider tail muscles, and shorter bodies. Mouth width was not significantly affected by competitors or predators.
When the four phenotypes were reared in the presence and absence of subsequent competition, they differed in their growth ability (Table 4, Fig. 4). The new
leopard frog competitors reduced growth of all four
wood frog phenotypes (F1,35 5 35.5, P , 0.00001). In
comparing growth among the four wood frog phenotypes, wood frogs previously reared in high competition environments grew faster than wood frogs previously reared in low competition environments (F1,35 5
21.2, P 5 0.00005). In contrast, wood frogs previously
reared in predator environments grew slower than wood
frogs previously reared in predator-free environments
(F1,35 5 10.6, P 5 0.003). There were no differences
in activity among the phenotypes due to past exposure
to either competitors (F1,16 5 0.1, P 5 0.751) or predators (F1,17 5 0.4, P 5 0.555).
Experiment 3: Identification of competitor cues
The third experiment compared the impact of manipulating food either by directly reducing food with
a constant density of tadpoles or by holding food constant and with increased tadpole densities. Per capita
food reduction and the method of per capita food reduction both had significant multivariate effects on
wood frog responses (Table 5, Fig. 5). Per capita food
reduction caused slower growth, higher activity, smaller muscle dimensions, and larger body dimensions.
While tadpole growth was not different between the
two competitive manipulations, tadpoles displayed
higher activity and longer tails when I manipulated
competitor density but deeper bodies when I manipulated food density.
The follow-up experiment examined wood frog activity at two water volumes and two densities (Fig. 6).
Immediately after the start of the experiment, tadpoles
were more active in shallow water (F1,12 5 8.9, P 5
0.011) but density had no effect (F1,12 5 0.1, P 5
0.756). Over the duration of the experiment, tadpoles
were still more active in shallow water (F1,12 5 5.0, P
5 0.045) and more active at high density ( F1,12 5 26.0,
P 5 0.0003).
DISCUSSION
Competition had widespread effects on the growth,
behavior, and morphology of larval wood frogs. In all
of the experiments, intraspecific and interspecific competitors decreased wood frog growth by similar
amounts, presumably by decreasing the level of resources available in the mesocosms. Larval anurans are
thought to be generalist herbivores that commonly experience reduced growth under high densities (Adolph
1931, Martof 1956, DeBenedictis 1974, Morin 1983,
Wilbur and Fauth 1990, Werner and Anholt 1996, Semlitsch et al. 1997, Werner and Glennemeier 1999, Relyea 2000; however, see Petranka and Kennedy 1999
for evidence of tadpole omnivory). In experiment 1,
we can directly compare the impact of intraspecific and
interspecific competition on wood frog tadpoles. As
November 2002
COMPETITOR-INDUCED TADPOLES
529
FIG. 1. Growth rate (6 1 SE), activity, and relative morphology of larval wood frogs (mean 6 1 SE) as a function of the
density of intraspecific and interspecific competitors. Differences in overall size were removed prior to analysis or morphological traits by regressing log-transformed linear measures against log-transformed mass and analyzing the residuals.
expected, conspecific competitors had a stronger effect
on wood frog growth than did the heterospecific competitors. The mechanism underlying this phenomenon
may be due to incomplete diet overlap between interspecific competitors or it may be that wood frogs have
an innate ability to grow faster than chorus frogs, lead-
ing to greater resource consumption and a greater intensity of competition.
Given that competitors were apparently reducing resources, both intraspecific and interspecific competition should have increased the activity of wood frog
tadpoles in experiment 1. Both theoretical models (Mc-
RICK A. RELYEA
530
TABLE 2. MANOVA results examining how wood frog survivorship, activity, and relative morphology were affected
by experimental block, wood frog population, competitor
intensity, and the species of competitor (intraspecific wood
frogs or interspecific chorus frogs; experiment 1).
A) Multivariate tests
Factor
Block
Population
Species
Competitive intensity
Population 3 Species
Population 3 Competitive
intensity
Species 3 Competitive intensity
df
F
P
10,25
30,74
10,25
10,25
30,74
30,74
5.39
1.63
0.49
6.46
1.41
0.89
0.0003
0.046
0.879
0.00007
0.116
0.630
10,25
2.23
0.051
B) Univariate tests
Factor
Survivorship
Activity
Tail length
Tail depth
Muscle depth
Muscle width
Body length
Body depth
Body width
Mouth width
Competitive
intensity
Species 3
Competitive
intensity
0.919
0.051
0.0001
0.017
,0.00001
,0.00001
0.002
0.0005
0.075
0.011
0.612
0.032
0.029
0.081
0.002
0.489
0.599
0.211
0.039
0.855
Notes: All three-way, four-way, and block interactions were
nonsignificant and pooled with the error term. Part (A) reports
the multivariate tests, and part (B) reports the univariate tests
for all significant multivariate tests.
Namara and Houston 1987, 1994, Houston et al. 1993,
Werner and Anholt 1993) and empirical studies (Akre
and Johnson 1979, Crowley 1979, Formanowicz 1982,
Inoue and Matsura 1983, Macchiusi and Baker 1992,
Horat and Semlitsch 1994, Anholt et al. 1996) support
the notion that decreased resources should induce an
increase in foraging activity. However, in my study,
activity increased only with intraspecific competitors.
Models of optimal activity implicitly assume that other
foraging-related traits (e.g., morphology) do not
change. This may be a reasonable assumption for shorter term experiments and for taxa that do not alter other
traits. However, we now know that these are not reasonable assumptions for longer term experiments on
larval anurans. It may be that morphological plasticity
can be used as a long-term alternative to behavioral
plasticity.
The morphological changes that occurred were striking: of the eight traits measured across three experiments and five competition treatments, four to seven
traits were significantly altered by competition. When
there was increased intraspecific competition, wood
frogs typically developed smaller tail dimensions and
larger body dimensions (although not all changes were
significant). These data are further supported by experiments that I have subsequently conducted on intraspecific competition in gray treefrog tadpoles (Hyla
versicolor; Relyea 2002b).
Ecological Monographs
Vol. 72, No. 4
In an earlier study of competitor-induced morphology in wood frogs, I documented small competitorinduced changes in morphology (Relyea 2000). In that
study, wood frogs did not significantly reduce growth
but they did develop relatively wider mouths and longer
tails compared to when reared alone. In the present
study, wood frog growth was reduced dramatically and
changes in wood frog morphology were more widespread among the measured traits, including the development of a wider mouth (experiment 1). However,
in the present study, tail length became shorter with
competition. The inconsistent tail length result across
experiments also has been observed in past studies of
predator-induced morphology; in separate experiments,
predators have induced increases, decreases, and no
change in tail length (Van Buskirk and Relyea 1998,
Relyea 2000, Relyea and Werner 2000). Thus, the tail
length response may also be an inconsistently changing
trait across competitor-induction experiments. A comparison of Relyea (2000) and the current study suggests
that weak competition significantly alters a small number of morphological traits but strong competition alters a large number of morphological traits.
The different magnitudes of response in the presence
of intraspecific and interspecific competition suggested
that wood frogs can discriminate among competitor
species. In experiment 1, wood frogs employed different strategies against intraspecific competition than
interspecific competition. When examined on a perunit-of-competition basis (i.e., competition intensity),
wood frogs responded to intraspecific competition by
increasing their activity and developing less extreme
changes in morphology. In contrast, they responded to
interspecific competition by maintaining constant activity but developing more extreme changes in morphology. Even more striking was the difference in response to interspecific competition from chorus frogs
in experiment 1 and interspecific competition from
leopard frogs in experiment 2. In both experiments,
wood frogs responded to the interspecific competition
by increasing their body length, but several of the other
morphological dimensions did not change in the same
direction. Experiment 3 demonstrated that wood frogs
can even discriminate between competition due to directly reduced food rations vs. competition due to increased conspecific density. Collectively, it appears
that wood frog tadpoles possess a keen ability to discriminate among different kinds of competition and
exhibit competitor-specific strategies.
The competitor-induced responses were continuous
responses rather than threshold responses (i.e., ‘‘on–
off switches’’). In experiments 1 and 3, further increases in competition led to more extreme behavioral and
morphological phenotypes. This suggests that tadpoles
are able to finely tune their phenotype to current competitive environments. In contrast to these competitorinduced responses, we do not yet know whether predator-induced responses of larval anurans are continuous
November 2002
COMPETITOR-INDUCED TADPOLES
531
FIG. 2. Activity and relative morphology of larval wood frogs (mean 6 1 SE) as a function of the competitive intensity
of intraspecific and interspecific competitors. Differences in overall size were removed prior to analysis by regressing logtransformed linear measures against log-transformed mass and analyzing the residuals.
responses or threshold responses (Van Buskirk and Relyea 1998, Relyea 2000, Relyea and Werner 2000). If
predator-induced responses also are continuous, then it
argues for an in-depth examination of how well larval
amphibians (as well as many other taxa) can finely tune
their phenotypes to different combinations of predator
and competitor environments.
There is a relatively small literature on competitorinduced morphological plasticity in animals. There are
a number of marine invertebrates including anemones,
corals, hydrozoans, and bryozoans that produce spines,
tentacles, and stolons in response to interference competitors (Stemberger and Gilbert 1987, Harvell 1990).
There are also a few larval amphibians that inhabit
RICK A. RELYEA
532
TABLE 3. MANOVA results examining how the growth rate,
activity, and relative morphology of wood frogs were affected by a factorial combination of competitors (present
and absent) and caged predators (present and absent; experiment 2, stage 1).
A) Multivariate tests
Factor
Competitor
Predator
Competitor 3
Predator
df
F
10,7
10,7
10,7
28.4
55.6
19.0
P
0.0001
0.00001
0.004
B) Univariate tests
Factor
Competitor
Growth rate
Activity
Tail length
Tail depth
Muscle depth
Muscle width
Body length
Body depth
Body width
Mouth width
,0.00001
,0.00001
0.002
0.685
0.413
0.288
0.002
0.007
0.001
0.326
Predator
0.127
,0.00001
0.0003
,0.00001
0.008
0.019
,0.00001
0.725
0.119
0.085
Competitor
3 Predator
0.500
,0.00001
0.0002
0.0005
0.836
0.038
0.214
0.612
0.618
0.510
highly ephemeral environments and can develop carnivorous and cannibalistic phenotypes under high competition (Bragg 1956, Pfennig 1992a,b, Reilly et al.
1992, Pfennig and Collins 1993). The cannibalistic responses are fundamentally different than the changes
observed in the wood frogs, which remain omnivorous.
While it is too early to know whether noncarnivorous,
competitor-induced morphological change is common
in larval anurans, subsequent experiments using a second species (Hyla versicolor) have demonstrated very
similar responses, suggesting that the phenomenon is
not restricted to wood frogs (Relyea 2002b). There also
are some animals, including grasshoppers (Melanus femurrubrum) and fish (Lepomis gibbosus, Cichlasoma
sp., and Geophagus spp.), that can alter their morphology when fed different diets (Meyer 1987, Wainwright et al. 1991, Wimberger 1991, Thompson 1992).
If these cases of diet-induced morphological plasticity
occur because competitors alter the type of food available, then these case studies also might be viewed as
competitor-induced morphological plasticity, suggesting that competitor-induced morphological plasticity
might be much more common than is currently appreciated.
Are competitor-induced changes adaptive?
For phenotypically plastic traits to evolve, it has been
hypothesized that no single phenotype can be optimal
in all environments (Via and Lande 1985, 1987, Via
1987, Gomulkiewicz and Kirkpatrick 1992). More specifically to this study, the theory predicts that competitor-induced traits should provide a fitness benefit
in a competitive environment but a fitness cost in an
alternative environment. With regard to behavior, it is
well established that increased activity results in great-
Ecological Monographs
Vol. 72, No. 4
er rates of harvesting resources and more rapid growth
in a variety of taxa including larval amphibians (Sih
1987, Lima and Dill 1990, Skelly 1992). By increasing
activity, individuals obtain a larger fraction of the
available resources and become a better ‘‘effect competitor’’ (sensu Goldberg 1990); this result has been
borne out in several recent experiments with larval anurans (Werner and Anholt 1996, Peacor and Werner
1997, Relyea and Werner 1999). Thus, the competitorinduced behavior observed in this study appears to be
adaptive.
With regard to morphology, both intraspecific and
interspecific competitors caused wood frog tadpoles to
develop relatively small tails and large bodies. Several
studies examining the costs and benefits of predatorinduced plasticity in tadpoles have shown that larvae
with relatively shallow tails and long bodies (the predator-free phenotype) grow more rapidly than larvae
with relatively large tails and small bodies (the predator-induced phenotype; Van Buskirk et al. 1997, Van
Buskirk and Relyea 1998, experiment 2 of the current
study). If these morphological traits are responsible for
the differences in growth rate, we would expect that
the competitor-induced wood frogs also should grow
more rapidly and be more competitive since they have
consistently longer bodies and often shallower tails.
Experiment 2 demonstrated that competitor-induced
tadpoles did experience higher growth than noninduced
tadpoles. Because there are differences in morphology
among the four phenotypes going into the performance
test but no differences in activity level, I can conclude
that the differences in subsequent growth among the
four phenotypes were not related to activity. This provides further evidence that changes in morphology can
have substantial impacts on larval anuran growth.
Growth rate is a relevant index of fitness in amphibians
because more rapid growth is positively related to several fitness-related life history parameters, including
the ability to metamorphose from a drying pond, survival in the terrestrial stage, and time to sexual maturity
(Berven and Gill 1983, Smith 1983, Newman 1988,
Semlitsch et al. 1988). In short, competitor-induced
morphological changes appear to be adaptive responses
to competition.
Whereas the competitor-induced phenotype experiences a growth benefit in the presence of competitors,
theory argues that this phenotype must suffer lower
fitness in an alternative environment. An excellent candidate is an environment containing predators. Past
studies investigating predator selection on tadpole morphology have repeatedly demonstrated that larval anurans with relatively shallow tails and large bodies (the
competitor-induced phenotype) are preferentially killed
by aquatic predators (McCollum and Van Buskirk
1996, Van Buskirk et al. 1997, Van Buskirk and Relyea
1998). The mechanism underlying this result appears
to be a reduction in the ability of shallow-tailed tadpoles to accelerate away from predators and avoid their
November 2002
COMPETITOR-INDUCED TADPOLES
533
FIG. 3. Relative morphology of larval wood frogs (mean 6 1 SE ) reared in experiment 2 under a factorial combination
of competition and caged predators (NP, no predator; P, predator; LC, low competition; HC, high competition). Differences
in overall size were removed prior to analysis by regressing log-transformed linear measures against log-transformed mass
and analyzing the residuals.
strikes (McCollum and Leimberger 1997, Relyea
2001c). Based on these studies, it appears that the fitness effects of competitor-induced morphology match
the pattern of competitor-induced behavior: it provides
more rapid growth in the presence of competitors but
lower survival in the presence of predators.
An alternative to the adaptive plasticity hypothesis
is that the competitor-induced changes in morphology
were simply the result of allometric growth. One could
argue that because competition slowed tadpole growth,
the differences in tadpole morphology between competitor treatments reflect the fact that smaller, less developed tadpoles normally exhibit relatively small tails
and large bodies. Previous work tracking the ontogeny
of morphology in larval wood frogs indicates that, in
fact, smaller, less developed tadpoles possess relatively
RICK A. RELYEA
534
TABLE 4. MANOVA results examining how the survivorship, growth rate, and activity of wood frogs were affected
by previous exposure to competitors or caged predators and
subsequent exposure to a second round of low or high
competition (experiment 2, stage 2).
A) Multivariate tests
Factor
Previous competitors
Previous predators
Subsequent competition
df
F
P
3,33
3,33
3,33
6.9
3.5
12.6
0.001
0.026
0.00001
B) Univariate tests
Factor
Survivorship
Growth rate
Activity
Previous
competitors
Previous
predators
Subsequent
competition
0.654
0.00005
0.414
0.654
0.003
0.726
0.575
,0.00001
0.119
Note: All interactions were nonsignificant and pooled with
the error term.
large tails and small bodies (Relyea and Werner 2000).
Thus, the morphological responses to competition were
not due to allometric effects.
What are the cues for competitor-induced plasticity?
The competitor cues that tadpoles use to alter their
growth and development have received a fair amount
of attention over the years and most studies have focused on the presence of chemical growth inhibitors
that may be related to the presence of a yeast ( Prototheca richardsi) in laboratory experiments (see reviews by Beebee 1995, Griffiths 1995, Petranka 1995).
However, as pointed out by Petranka (1995), the yeast
are not well correlated in field conditions and there are
a number of other possible chemical cues that have yet
to be investigated.
Less attention has been paid to the competitor cues
that affect tadpole behavior and morphology. There
have been a number of studies (in a wide range of taxa)
that document how prey alter their activity in response
to either higher densities of competitors or lower food
resources (Anholt and Davies 1987, Macchiusi and
Baker 1992, Anholt and Werner 1995, Van Buskirk and
Yurewicz 1998). However, there have been no direct
comparisons of the two mechanisms until the current
study. The most parsimonious hypothesis is that individuals simply respond to decreased resources. An alternative hypothesis is that tadpoles respond to either
visual or chemical cues (perhaps similar to the cues
emitted by predators; Petranka et al. 1987, Kats et al.
1988, McCollum and Leimberger 1997, Kats and Dill
1998).
Experiment 3 allowed me to address these two hypotheses (decreased food vs. increased density). In this
experiment, wood frog growth rates were nearly identical when I altered competition either by increasing
density or decreasing resources; this demonstrated that
both treatments had the same competitive impact on
tadpole performance (for different growth outcomes,
Ecological Monographs
Vol. 72, No. 4
see Wilbur 1977, Murray 1990). Of the eight behavioral
and morphological traits, five were altered similarly
regardless of ration method. However, three of the traits
responded differently to the two ration methods. For
example, wood frog tadpoles increased their activity
by 10% with decreased food rations but they increased
their activity by 23% with increased conspecific density. The fact that the two ration methods ultimately
produced the same growth rate but different phenotypes
suggests that wood frogs can discriminate between
competition due to reduced food and competition due
to increased density. It also suggests that wood frogs
used unique phenotypic strategies to arrive at the same
growth rate. These results mirror those found in experiment 1 in which wood frogs exhibited different
behavioral and morphological responses to the same
competitive intensities of intra- and interspecific competitors. Collectively, these data demonstrate that tadpoles use both reduced food and increased density to
assess the level of competition in their environment.
Tadpoles could assess increased densities (independent of food) through visual, tactile, or chemical cues.
At the start of the follow-up experiment using different
water depths (before food limitation had a chance to
occur) tadpole density did not affect activity. Because
visual and tactile cues would be immediately apparent
to tadpoles, I rejected the hypothesis that tadpoles use
visual or tactile cues to assess density. Further, visual
and tactile cues should have been the same for eight
tadpoles reared in 6-L tubs and 16 tadpoles reared in
12-L tubs; however, activity differed between these two
treatments, further rejecting the visual and tactile hypotheses. This leaves the possibility that tadpoles assess density (independent of food) through chemical
cues. In contrast to visual and tactile cues, chemical
cues (e.g., feces or urine) probably take several hours
or days to accumulate and would not be immediately
apparent to the tadpoles. Identification of these chemical cues should prove to be an intriguing area of future
research.
Linking competitor- and predator-induced plasticity
The competitor-induced responses and fitness effects
detailed in this study of wood frog larvae are in direct
contrast to predator-induced responses and fitness effects in wood frog larvae (Van Buskirk and Relyea
1998, Relyea 2000, 2001a,b, 2002a, Relyea and Werner
2000) and other species of larval anurans (Smith and
Van Buskirk 1995, McCollum and Van Buskirk 1996,
McCollum and Leimberger 1997, Van Buskirk et al.
1997, Van Buskirk and Relyea 1998). Many predators
induce low activity, deep tails, and short bodies whereas competitors in the present study generally induced
high activity, shallow tails, and long bodies. The predator-induced suite of traits makes prey more resistant
to predators but at a cost of reduced growth and development (McCollum and Van Buskirk 1996, Van
November 2002
COMPETITOR-INDUCED TADPOLES
535
FIG. 4. Larval wood frog growth in experiment 2 (mean 6 1 SE). The upper panel reflects wood frog growth after the
initial exposure to the factorial combination of competition and caged predators. The bottom panel reflects relative wood
frog growth in the subsequent performance trial with and without additional competition. Relative growth was calculated as
(final mass)/(mass at the beginning of the performance trial). Abbreviations are as in Fig. 3.
Buskirk et al. 1997, Van Buskirk and Relyea 1998, see
also Results). In contrast, the competitor-induced suite
of traits makes tadpoles more competitive, but these
same traits make tadpoles more vulnerable to predators.
Thus, it appears that competitive ability and predator
resistance ability are being traded off in phenotypically
plastic larval anurans.
This trade-off may also exist in many other taxa. The
trade-off certainly exists for the numerous taxa that
alter their behavior in response to predators and competitors (see reviews by Havel 1987, Lima and Dill
1990, Werner and Anholt 1993, Kats and Dill 1998).
In contrast to the plethora of behavioral studies, we
have a small but growing number of studies on predator-induced morphology and a paucity of data on competitor-induced morphology. The lack of competitorinduced studies is likely more due to a lack of inves-
tigation into the phenomenon rather than a lack of its
occurrence. There is at least one other study system
that has suggestive data of a trade-off between competitor- and predator-induced morphology. The crucian
carp (Carassius carassius) has been well characterized
for its ability to develop a deep-bodied phenotype in
the presence of predatory pike (Esox lucius), which
makes it less vulnerable to predation but at a competitive cost (Pettersson and Brönmark 1997). Other investigators have noted that the carp exhibits a shallowbodied phenotype in lakes with high densities of conspecifics and these shallow-bodied fish become deeper
bodied with either the addition of predator cues or the
addition of high food resources (Holopainen et al.
1997). Much like the initial work on the larval anuran
system, the carp system has been treated primarily in
terms of predator-induced and noninduced prey. It may,
536
RICK A. RELYEA
Ecological Monographs
Vol. 72, No. 4
FIG. 5. The growth, activity, and relative morphology of larval wood frogs (mean 6 1 SE) reared at different levels of
intraspecific competition, either directly by reduction of food rations to a constant density of tadpoles or indirectly by
increasing tadpole density on a constant food ration. Differences in overall size were removed prior to analysis by regressing
log-transformed linear measures against log-transformed mass and analyzing the residuals.
in fact, be another excellent example of competitorand predator-induced plasticity interacting together
over ecological and evolutionary time.
A trade-off between competitive ability and predator
resistance makes sense in light of the range of environments that many prey experience in nature. For ex-
ample, in natural ponds, wood frogs live along a continuum of predator and competitor environments. In
ponds with high densities of predators, we typically
observe very low densities of wood frogs (1–5 frogs/
m2; E. E. Werner, R. A. Relyea, D. K. Skelly, and K.
L. Yurewicz, unpublished data). At such low densities,
COMPETITOR-INDUCED TADPOLES
November 2002
TABLE 5. MANOVA results examining how wood frog tadpoles alter their growth, activity, and morphology when
they are reared at different levels of intraspecific competition (per capita food rations) using two different ration
methods, either directly by reduction of food rations to a
constant density of tadpoles or indirectly by increasing tadpole density on a constant food ration.
A) Multivariate tests
Factor
Per capita ration
Ration method
Per capita ration 3 Ration method
df
F
P
10,11
10,11
10,11
12.8
4.4
2.2
0.001
0.012
0.106
B) Univariate tests
Factor
Per capita
ration
Ration
method
Per capita
ration 3
Ration
method
Growth rate
Activity
Tail length
Tail depth
Muscle depth
Muscle width
Body length
Body depth
Body width
Mouth width
0.00001
,0.00001
0.318
0.401
0.058
0.013
0.015
0.088
0.010
0.599
0.403
0.00008
0.022
0.445
0.933
0.369
0.539
0.012
0.963
0.670
0.693
0.003
0.039
0.737
0.956
0.615
0.648
0.051
0.665
0.996
competition for resources might be minimal, although
behavioral responses might cause prey to congregate
into refuges and experience higher competition. In contrast, in ponds that contain few or no predators, we
typically observe very high densities of wood frogs (up
to 450 frogs/m2; E. E. Werner et al., unpublished data).
At such high densities, competition for resources
should be much more intense. Thus, wood frogs (and
probably many other taxa) experience a range of environments from high predation/low competition to low
predation/high competition and these alternative environments would select for alternative suites of traits.
A competition–predation trade-off across species has
become a paradigm in community ecology because it
can result in a wide range of numerically mediated
indirect effects, including keystone predation, indirect
mutualisms, and trophic cascades (Hairston et al. 1960,
Brooks and Dodson 1965, Paine 1966, Dodson 1970,
Levine 1976, Lubchenco 1978, Vandermeer 1980, Morin 1981, Menge et al. 1994, Werner and McPeek 1994,
Wootton 1994). Indeed, the species-level trade-off between competitive ability and predation resistance has
become such a strong paradigm that many theoretical
models of species interactions use this trade-off as an
underlying assumption (Vance 1978, Abrams 1982,
1984, 1995, Werner and Anholt 1993, Holt and Lawton
1994, Leibold 1996). With the results of the current
study, it now appears that this paradigm not only applies across species, but also may apply to alternative
phenotypes expressed within species.
537
CONCLUSIONS
This study suggests that we should strive to examine
connections between seemingly disparate forms of phenotypic plasticity. Predator- and competitor-induced
plasticity have long been treated as two distinct areas
of study but they may be a single interrelated phenomenon. Organisms experience a wide range of biotic and
abiotic environments and these environments often are
related to each other (e.g., low predation/high competition environments). Therefore, many types of plasticity should be related to each other and interact with
each other over ecological and evolutionary time. By
making these connections, we will obtain new insights
into the evolution of plastic traits and we will likely
find within-species support for cross-species paradigms. By focusing on plastic traits, we also will improve our understanding of trait function and the mechanisms underlying ecological paradigms. In summary,
by considering the inter-relatedness of adaptive responses, we stand to gain insights into the evolution
of traits and the mechanisms of ecological interactions
that shape natural communities.
FIG. 6. The impact of tadpole density and water volume
on the activity of wood frog tadpoles (mean 6 1 SE). The top
panel represents the activity of tadpoles after 3-h exposure
to the treatments. The bottom panel represents the activity of
tadpoles over the entire 6-d duration of the experiment.
RICK A. RELYEA
538
ACKNOWLEDGMENTS
I thank J. Hoverman, J. Moll, and K. Wittkopp for assistance with the field work and R. Nussbaum and R. Alexander
for providing access to the E. S. George Reserve. K. Wittkopp
and C. Glaude did outstanding jobs digitizing the tadpoles.
S. Ball, E. Werner, and K. Yurewicz provided helpful comments on the manuscript. This work was supported by University of Michigan research grants, Sigma Xi research grants,
and National Science Foundation grants DEB-9701111 and
DEB-9903761.
LITERATURE CITED
Abrams, P. A. 1982. Functional responses of optimal foragers. American Naturalist 120:382–390.
Abrams, P. A. 1984. Foraging time optimization and interactions in food webs. American Naturalist 124:80–96.
Abrams, P. A. 1995. Implications of dynamically variable
traits for identifying, classifying, and measuring direct and
indirect effects in ecological communities. American Naturalist 146:112–134.
Adolph, E. F. 1931. The size of the body and the size of the
environment in the growth of tadpoles. Biological Bulletin
61:350–375.
Akre, B. G., AND D. M. Johnson. 1979. Switching and sigmoid functional response curves by damselfly naiads with
alternate prey available. Journal of Animal Ecology 48:
703–720.
Altmann, J. 1974. Observational study of behavior: sampling
methods. Behaviour 49:227–267.
Anholt, B. R., and R. W. Davies. 1987. Effect of hunger level
on the activity of the predatory leech Nephelopsis obscura
Verrill (Hirudinoidea: Erpobdellidae). American Midland
Naturalist 117:307–311.
Anholt, B. R., D. K. Skelly, and E. E. Werner. 1996. Factors
modifying antipredator behavior in larval toads. Herpetologica 52:301–313.
Anholt, B. R., and E. E. Werner. 1995. Interaction between
food availability and predation mortality mediated by adaptive behavior. Ecology 76:2230–2234.
Barrett, S. C. H., and B. F. Wilson. 1981. Colonizing ability
in the Echinochloa crus-galli complex (barnyard grass). I.
Variation in life history. Canadian Journal of Botany 59:
1844–1860.
Beebee, T. J. C. 1995. Tadpole growth: is there an interference
effect in nature? Herpetological Journal 5:204–205.
Berven, K. A., and D. E. Gill. 1983. Interpreting geographic
variation in life-history traits. American Zoologist 23:85–
97.
Bookstein, F. L. 1989. ‘‘Size and shape’’: a comment on
semantics. Systematic Zoology 38:173–180.
Bradshaw, A. D. 1965. Evolutionary significance of phenotypic plasticity in plants. Advances in Genetics 13:115–
155.
Bragg, A. N. 1956. Dimorphism and cannibalism in tadpoles
of Scaphiopus bombifrons (Amphibia, Salientia). Southwestern Naturalist 1:105–108.
Brooks, J. L., and S. I. Dodson. 1965. Predation, body size,
and composition of plankton. Science 150:28–35.
Collins, J. P., and H. M. Wilbur. 1979. Breeding habits and
habitats of the amphibians of the Edwin S. George Reserve,
Michigan, with notes on the local distribution of fishes.
Occasional Papers of the Museum of Zoology, University
of Michigan. Number 686.
Crowley, P. H. 1979. Behavior of zygopteran nymphs in a
simulated weed bed. Odonatologica 8:91–101.
DeBenedictis, P. A. 1974. Interspecific competition between
tadpoles of Rana pipiens and Rana sylvatica: an experimental field study. Ecological Monographs 44:129–151.
Dodson, S. 1989. Predator-induced reaction norms. BioScience 39:447–452.
Ecological Monographs
Vol. 72, No. 4
Dodson, S. I. 1970. Complementary feeding niches sustained
by size-selective predation. Limnology and Oceanography
15:131–137.
Dudley, S. A., and J. Schmitt. 1996. Testing the adaptive
plasticity hypothesis: density-dependent selection on manipulated stem length in Impatiens capensis. American Naturalist 147:445–465.
Formanowicz, D. R., Jr. 1982. Foraging tactics of larvae of
Dytiscus verticalis (Coleoptera: Dytiscidae): the assessment
of prey density. Journal of Animal Ecology 51:757–767.
Goldberg, D. E. 1990. Components of resource competition
in plant communities. Pages 27–49 in J. Grace and D. Tilman, editors. Perspectives in plant competition. Academic
Press, San Diego, California, USA.
Gomulkiewicz, R., and M. Kirkpatrick. 1992. Quantitative
genetics and the evolution of reaction norms. Evolution 46:
390–411.
Gosner, K. L. 1960. A simplified table for staging anuran
embryos and larvae with notes on identification. Herpetologica 16:183–190.
Griffiths, R. A. 1995. Determining competition mechanisms
in tadpole assemblages. Herpetological Journal 5:208–210.
Hairston, N. G., F. E. Smith, and L. B. Slobodkin. 1960.
Community structure, population control, and competition.
American Naturalist 94:421–424.
Harvell, C. D. 1990. The ecology and evolution of inducible
defenses. Quarterly Review of Biology 65:323–340.
Havel, J. E. 1987. Predator-induced defenses: a review. Pages
263–278 in W. C. Kerfoot and A. Sih, editors. Predation:
direct and indirect impacts on aquatic communities. University Press of New England, Hanover, New Hampshire,
USA.
Holopainen, I. J., J. Aho, M. Vornanen, and H. Huuskonen.
1997. Phenotypic plasticity and predator effects on morphology and physiology of crucian carp in nature and in
the laboratory. Journal of Fish Biology 50:781–798.
Holt, R. D., and J. H. Lawton. 1994. The ecological consequences of shared natural enemies. Annual Review of
Ecology and Systematics 25:495–520.
Horat, P., and R. D. Semlitsch. 1994. Effects of predation
risk and hunger on the behaviour of two species of tadpoles.
Behavioral Ecology and Sociobiology 34:393–401.
Houston, A. I., J. M. McNamara, and J. M. C. Hutchinson.
1993. General results concerning the trade-off between
gaining energy and avoiding predation. Philosophical
Transactions of the Royal Society of London Series B 341:
375–397.
Inoue, T., and T. Matsura. 1983. Foraging strategy of a mantid, Paratenodera angustipennis S.: mechanisms of switching tactics between ambush and active search. Oecologia
56:264–271.
Karban, R., and I. T. Baldwin. 1997. Induced responses to
herbivory. University of Chicago Press, Chicago, Illinois,
USA.
Kats, L. B., and L. M. Dill. 1998. The scent of death: chemosensory assessment of predation risk by prey animals.
Ecoscience 5:361–394.
Kats, L. B., J. W. Petranka, and A. Sih. 1988. Antipredator
defenses and the persistence of amphibian larvae with fishes. Ecology 69:1865–1870.
Leibold, M. A. 1996. A graphical model of keystone predator
in food webs: trophic regulation of abundance, incidence,
and diversity patterns in communities. American Naturalist
147:785–814.
Levine, S. H. 1976. Competitive interactions in ecosystems.
American Naturalist 110:903–910.
Lima, S. L., and L. M. Dill. 1990. Behavioral decisions made
under the risk of predation: a review and prospectus. Canadian Journal of Zoology 68:619–640.
Lubchenco, J. 1978. Plant species diversity in a marine in-
November 2002
COMPETITOR-INDUCED TADPOLES
tertidal community: importance of herbivore food preference and algal competitive abilities. American Naturalist
112:23–39.
Macchiusi, F., and R. L. Baker. 1992. Effects of predators
and food availability on activity and growth of Chironomus
tentans (Chironomidae: Diptera). Freshwater Ecology 28:
207–216.
Martof, B. 1956. Factors influencing size and composition
of populations of Rana clamitans. American Midland Naturalist 56:224–245.
McCollum, S. A., and J. D. Leimberger. 1997. Predator-induced morphological changes in an amphibian: predation
by dragonflies affects tadpole shape and color. Oecologia
109:615–621.
McCollum, S. A., and J. Van Buskirk. 1996. Costs and benefits of a predator-induced polyphenism in the gray tree
frog Hyla chrysocelis. Evolution 50:583–593.
McNamara, J. M., and A. I. Houston. 1987. Starvation and
predation as factors limiting population size. Ecology 68:
1515–1519.
McNamara, J. M., and A. I. Houston. 1994. The effect of a
change in foraging options on intake rate and predation
rate. American Naturalist 144:978–1000.
Menge, B. A., E. L. Berlow, C. A. Blanchette, S. A. Navarrette, and S. B. Yamada. 1994. The keystone species concept: variation in interaction strength in a rocky intertidal
habitat. Ecological Monographs 64:249–286.
Meyer, A. 1987. Phenotypic plasticity and heterochrony in
Cichlasoma managuense (Pisces, Cichlidae) and their implications for speciation in cichlid fishes. Evolution 41:
1357–1369.
Morin, P. J. 1981. Predatory salamanders reverse the outcome
of competition among three species of anuran tadpoles.
Science 212:1284–1286.
Morin, P. J. 1983. Predation, competition, and the composition of larval anuran guilds. Ecological Monographs 53:
119–138.
Murray, D. L. 1990. The effects of food and density on
growth and metamorphosis in larval wood frogs (Rana sylvatica) from central Labrador. Canadian Journal of Zoology
68:1221–1226.
Newman, R. A. 1988. Adaptive plasticity in development of
Scaphiopus couchii tadpoles in desert ponds. Evolution 42:
774–783.
Paine, R. T. 1966. Food web complexity and species diversity. American Naturalist 100:65–75.
Peacor, S. D., and E. E. Werner. 1997. Trait-mediated indirect
interactions in a simple aquatic food web. Ecology 78:
1146–1156.
Petranka, J. W. 1995. Interference competition in tadpoles:
are multiple agents involved? Herpetological Journal 5:
206–207.
Petranka, J. W., L. B. Kats, and A. Sih. 1987. Predator–prey
interactions among fish and larval amphibians: use of
chemical cues to detect predatory fish. Animal Behaviour
35:420–425.
Petranka, J. W., and C. A. Kennedy. 1999. Pond tadpoles
with generalized morphology: is it time to reconsider their
functional roles in aquatic communities. Oecologia 120:
621–631.
Pettersson, L. B., and C. Brönmark. 1997. Density-dependent
costs of an inducible morphological defense in crucian carp.
Ecology 78:1805–1815.
Pfennig, D. W. 1992a. Polyphenism in spadefoot toad tadpoles as a locally adjusted evolutionary stable strategy.
Evolution 46:1408–1420.
Pfennig, D. W. 1992b. Proximate and functional causes of
polyphenism in an anuran tadpole. Functional Ecology 6:
167–174.
539
Pfennig, D. W., and J. P. Collins. 1993. Kinship affects morphogenesis of cannibalistic tadpoles. Nature 362:836–838.
Reilly, S. M., G. V. Lauder, and J. P. Collins. 1992. Performance consequences of a trophic polymorphism: feeding
behavior in typical and cannibal phenotypes of Ambystoma
tigrinum. Copeia 1992:672–679.
Relyea, R. A. 2000. Trait-mediated effects in larval anurans:
reversing competitive outcomes with the threat of predation. Ecology 81:2278–2289.
Relyea, R. A. 2001a. Morphological and behavioral plasticity
of larval anurans in response to different predators. Ecology
82:523–540.
Relyea, R. A. 2001b. The lasting effects of adaptive plasticity: predator-induced tadpoles become long-legged frogs.
Ecology 82:1947–1955.
Relyea, R. A. 2001c. The relationship between predation risk
and antipredator responses in larval anurans. Ecology 82:
541–554.
Relyea, R. A. 2002a. Local population differences in phenotypic plasticity: predator-induced changes in wood frog
tadpoles. Ecological Monographs 72:77–93.
Relyea, R. A. 2002b. The many faces of predation: how
induction, selection, and thinning combine to alter prey
phenotypes. Ecology 83:1953–1964.
Relyea, R. A., and E. E. Werner. 1999. Quantifying the relation between predator-induced behavior and growth performance in larval anurans. Ecology 80:2117–2124.
Relyea, R. A., and E. E. Werner. 2000. Morphological plasticity of four larval anurans distributed along an environmental gradient. Copeia 2000:178–190.
Schlichting, C. D. 1986. The evolution of phenotypic plasticity in plants. Annual Review of Ecology and Systematics
17:667–693.
Schlichting, C. D., and D. A. Levin. 1984. Phenotypic plasticity of annual Phlox: tests of some hypotheses. American
Journal of Botany 71:252–260.
Schlichting, C. D., and M. Pigliucci. 1998. Phenotypic evolution: a reaction norm perspective. Sinauer, Sunderland,
Massachusetts, USA.
Semlitsch, R. D., H. Hotz, and G. Guex. 1997. Competition
among tadpoles of coexisting hemiclones of hybridogenetic
Rana esculenta: support for the frozen niche variation model. Evolution 51:1249–1261.
Semlitsch, R. D., D. C. Scott, and J. H. K. Pechmann. 1988.
Time and size at metamorphosis related to adult fitness in
Ambystoma talpoideum. Ecology 69:184–192.
Sih, A. 1987. Predators and prey lifestyles: an evolutionary
and ecological overview. Pages 203–224 in W. C. Kerfoot
and A. Sih, editors. Predation: direct and indirect impacts
on aquatic communities. University Press of New England,
Hanover, New Hampshire, USA.
Skelly, D. K. 1992. Field evidence for a cost of behavioral
antipredator response in a larval amphibian. Ecology 73:
704–708.
Smith, D. C. 1983. Factors controlling tadpole populations
of the chorus frog (Pseudacris triseriata) on Isle Royale,
Michigan. Ecology 64:501–510.
Smith, D. C., and J. Van Buskirk. 1995. Phenotypic design,
plasticity, and ecological performance in two tadpole species. American Naturalist 145:211–233.
Stemberger, R. S., and J. J. Gilbert. 1987. Multiple-species
induction of morphological defenses in the rotifer Keratella
testudo. Ecology 68:370–378.
Stephens, D. W., and J. R. Krebs. 1986. Foraging theory.
Princeton University Press, Princeton, New Jersey, USA.
Sultan, S. E. 1987. Evolutionary implications of phenotypic
plasticity in plants. Evolutionary Biology 21:127–178.
SYSTAT. 1992. Statistics version 5. 2 edition. SYSTAT, Evanston, Illinois, USA.
Thompson, D. B. 1992. Consumption rates and the evolution
540
RICK A. RELYEA
of diet-induced plasticity in the head morphology of Melanus femurrubrum (Orthoptera: Acrididae). Oecologia 89:
204–213.
Tollrian, R., and D. Harvell. 1999. The ecology and evolution
of inducible defenses. Princeton University Press, Princeton, New Jersey, USA.
Van Buskirk, J., S. A. McCollum, and E. E. Werner. 1997.
Natural selection for environmentally-induced phenotypes
in tadpoles. Evolution 52:1983–1992.
Van Buskirk, J., and R. A. Relyea. 1998. Natural selection
for phenotypic plasticity: predator-induced morphological
responses in tadpoles. Biological Journal of the Linnean
Society 65:301–328.
Van Buskirk, J., and K. L. Yurewicz. 1998. Effects of predators on prey growth rate: relative contributions of thinning
and reduced activity. Oikos 82:20–28.
Vance, R. 1978. Predation and resource partitioning in one
predator–two prey systems. American Naturalist 112:797–
813.
Vandermeer, J. 1980. Indirect mutualism: variation on a
theme by Stephen Levine. American Naturalist 116:441–
448.
Van Tienderen, P. H. 1991. Evolution of generalists and specialists in spatially heterogeneous environments. Evolution
456:1317–1331.
Via, S. 1987. Genetic constraints on the evolution of phenotypic plasticity. Pages 47–71 in V. Loeschke, editor. Genetic constraints on adaptive evolution. Springer, Berlin,
Germany.
Via, S., and R. Lande. 1985. Genotype–environment interaction and the evolution of phenotypic plasticity. Evolution
39:502–522.
Via, S., and R. Lande. 1987. Evolution of genetic variability
Ecological Monographs
Vol. 72, No. 4
in a spatially heterogeneous environment: effects of genotype–environment interaction. Genetic Research 49:
147–156.
Wainwright, P. C., C. W. Osenberg, and G. G. Mittlebach.
1991. Trophic polymorphism in the pumpkinseed sunfish
(Lepomis gibbosus Linnaeus): effects of environment on
ontogeny. Functional Ecology 5:40–55.
Werner, E. E., and B. R. Anholt. 1993. Ecological consequences of the trade-off between growth and mortality rates
mediated by foraging activity. American Naturalist 142:
242–272.
Werner, E. E., and B. R. Anholt. 1996. Predator-induced behavioral indirect effects in anuran larvae. Ecology 77:157–
169.
Werner, E. E., and K. S. Glennemeier. 1999. The influence
of forest canopy cover on the breeding pond distribution
of several amphibian species. Copeia 1999:1–12.
Werner, E. E., and M. A. McPeek. 1994. Direct and indirect
effects of predators on two anuran species along an environmental gradient. Ecology 75:1368–1382.
West-Eberhard, M. J. 1989. Phenotypic plasticity and the
origins of diversity. Annual Review of Ecology and Systematics 20:249–278.
Wilbur, H. M. 1977. Interactions of food level and population
density in Rana sylvatica. Ecology 58:206–209.
Wilbur, H. M., and J. E. Fauth. 1990. Experimental aquatic
food webs: interactions between two predators and two
prey. American Naturalist 135:176–204.
Wimberger, P. H. 1991. Plasticity of jaw and skull morphology in the Neotropical cichlids Geophagus brasiliensis
and G. Steindachneri. Evolution 45:1545–1563.
Wootton, J. T. 1994. Putting the pieces together: testing the
independence of interactions among organisms. Ecology
75:1544–1551.