Divergent walleye (Sander vitreus)-mediated inducible defenses in the centrarchid pumpkinseed sunfish (Lepomis gibbosus)

Predation has important ecological and evolutionary consequences. Evolutionary responses to diversifying selection include genetic differentiation, the evolution of adaptive phenotypic plasticity, and the genetic differentiation of plastic responses between populations. We tested if pumpkinseed sunfish ( Lepomis gibbosus ) respond to predation cues by changing their external body form in functionally sensible ways. We then asked whether predation has influenced the divergence of coexisting littoral and pelagic ecomorphs, by testing for divergent predator-induced responses. Juvenile L. gibbosus of both ecomorphs were reared with and without predation cues supplied by walleye ( Sander vitreus ) feeding on L. gibbosus . Predation cues stimulated increased body depth and dorsal spine length, but no increase in anal spine length or pectoral fin size. The dorsal spines of pelagic ecomorphs also grew longer than did those of littoral ecomorphs, while positive body depth responses were similar in both ecomorphs. This is the second fish taxa in which predator-induced morphological responses have been found, and the first in which divergent responses have been detected between ecomorphs. This suggests that the developmental systems of L. gibbosus ecomorphs have diverged under selection related to predation. We propose that other 'resource polymorphisms' in fishes have evolved under selection arising from a variety of factors, including predation, and not just selection related to resource use.  © 2007 The Linnean Society of London, Biological Journal of the Linnean Society , 2007, 90 , 25–36.     

Analyzing nested variation in the body form of Lepomid sunfishes

Phylogenetic hierarchies are often composed of younger diverging lineages nested within older diverging lineages. Comparing phenotypic variation among several hierarchical levels can be used to test hypotheses about selection, phenotypic evolution and speciation. Such hierarchical comparisons have only been performed in threespine stickleback, and so here we use a hierarchical pattern of divergences between near-shore littoral and off-shore pelagic habitats to test for selection on the evolution of body form in Lepomis sunfish in lakes. We compare variation in external body form between fish from littoral and pelagic habitats at three levels: among ecomorphs within individual lake populations (intrapopulation), among populations of the same species in different lakes (interpopulation), and between bluegill and pumpkinseed sunfish species (interspecifically). Using geometric morphometric methods, we first demonstrate that interpopulation variation in mean body form of pumpkinseed sunfish varies with the presence of pelagic habitat. We then incorporate these results with existing data in order to test the similarity of phenotypic divergence between littoral and pelagic habitats at different hierarchical levels. Parallel relationships between certain body form traits (head length, caudal length and pectoral length) and habitat occur at all three levels suggesting that selection persistently acts at all levels to diversify these traits and so may contribute to species formation. For other traits (caudal depth and pectoral altitude), divergence between habitats is inconsistent at different hierarchical levels. Thus, nested biological variation in Lepomid body form reflects a history of deterministic selection and historical contingency, and also identifies traits that likely have likely influenced fitness and so serve important functions.     

Assortative mating but no evidence of genetic divergence in a species characterized by a trophic polymorphism

Disruptive selection is a process that can result in multiple subgroups within a population, which is referred to as diversification. Foraging‐related diversification has been described in many taxa, but many questions remain about the contribution of such diversification to reproductive isolation and potentially sympatric speciation. Here, we use stable isotope analysis of diet and morphological analysis of body shape to examine phenotypic divergence between littoral and pelagic foraging ecomorphs in a population of pumpkinseed sunfish (Lepomis gibbosus). We then examine reproductive isolation between ecomorphs by comparing the isotopic compositions of nesting males to eggs from their nests (a proxy for maternal diet) and use nine microsatellite loci to examine genetic divergence between ecomorphs. Our data support the presence of distinct foraging ecomorphs in this population and indicate that there is significant positive assortative mating based on diet. We did not find evidence of genetic divergence between ecomorphs, however, indicating that isolation is either relatively recent or is not strong enough to result in genetic divergence at the microsatellite loci. Based on our findings, pumpkinseed sunfish represent a system in which to further explore the mechanisms by which natural and sexual selection contribute to diversification, prior to the occurrence of sympatric speciation.     

Foraging performance of diet-induced morphotypes in pumpkinseed sunfish (Lepomis gibbosus) favours resource polymorphism

Morphological plasticity can influence adaptive divergence when it affects fitness components such as foraging performance. We induced morphological variation in pumpkinseed sunfish (Lepomis gibbosus) ecomorphs and tested for effects on foraging performance. Young-of-year pumpkinseed sunfish from littoral and pelagic lake habitats were reared each on a 'specialist diet' representing their native habitat-specific prey, or a 'generalist diet' reflecting a combination of native and non-native prey. Specialist and generalist diets, respectively, induced divergent and intermediate body forms. Specialists had the highest capture success on their native prey whereas generalist forms were inferior. Specialists faced trade-offs across prey types. However, pelagic specialists also had the highest intake rate on both prey types suggesting that foraging trade-offs are relaxed when prey are abundant. This increases the likelihood of a resource polymorphism because the specialized pelagic form can be favoured by directional selection when prey are abundant and by diversifying selection when prey resources are restricted.     

Contemporary and historical evolutionary processes interact to shape patterns of within-lake phenotypic divergences in polyphenic pumpkinseed sunfish, Lepomis gibbosus

Historical and contemporary evolutionary processes can both contribute to patterns of phenotypic variation among populations of a species. Recent studies are revealing how interactions between historical and contemporary processes better explain observed patterns of phenotypic divergence than either process alone. Here, we investigate the roles of evolutionary history and adaptation to current environmental conditions in structuring phenotypic variation among polyphenic populations of sunfish inhabiting 12 postglacial lakes in eastern North America. The pumpkinseed sunfish polyphenism includes sympatric ecomorphs specialized for littoral or pelagic lake habitats. First, we use population genetic methods to test the evolutionary independence of within-lake phenotypic divergences of ecomorphs and to describe patterns of genetic structure among lake populations that clustered into three geographical groupings. We then used multivariate analysis of covariance (MANCOVA) to partition body shape variation (quantified with geometric morphometrics) among the effects of evolutionary history (reflecting phenotypic variation among genetic clusters), the shared phenotypic response of all populations to alternate habitats within lakes (reflecting adaptation to contemporary conditions), and unique phenotypic responses to habitats within lakes nested within genetic clusters. All effects had a significant influence on body form, but the effects of history and the interaction between history and contemporary habitat were larger than contemporary processes in structuring phenotypic variation. This highlights how divergence can be better understood against a known backdrop of evolutionary history.     

Does predation risk influence habitat use by northern redbelly dace Phoxinus eos at different spatial scales?

This study investigated the relationship between spatial variations in predation risk and abundance of northern redbelly dace Phoxinus eos at both macroscale (littoral v. pelagic zones) and microscale (structured v. open water habitats in the littoral zone) of Canadian Shield lakes. Minnow traps were placed in both structured and open water habitats in the littoral zone of 13 Canadian Shield lakes, and estimates of the relative predation risk of P. eos in both the pelagic and the littoral zones were obtained from tethering experiments. Results showed that (1) the mean abundance of P. eos in the littoral zone was positively correlated with the relative predation risk in the pelagic zone, (2) P. eos preferentially used structured over open water habitats in the littoral zone and (3) this preference was not related to the relative predation risk in the littoral zone but decreased as the relative predation risk increased in the pelagic zone. At the lake level, these results support the hypothesis that P. eos enter the littoral zone to avoid pelagic piscivores. At the littoral zone level, the results do not necessarily contradict the widely accepted view that P. eos preferentially use structured over open habitats to reduce their predation risk, but suggest that flexibility in antipredator tactics ( e.g. shelter use v. shoaling) could explain the spatial distribution of P. eos between structured and open water habitats.     

Morphological differentiation in introduced pumpkinseed Lepomis gibbosus (L.) occupying different habitat zones in Portuguese reservoirs

Introduced pumpkinseed Lepomis gibbosus sampled from four habitat zones (fluvial pelagic, fluvial littoral, lacustrine pelagic and lacustrine littoral) in three Portuguese reservoirs were used to test the hypotheses that habitats with the least similar characteristics will show the most differentiation, and that morphological differences will relate to functional adaptations to flow and trophic habitats. Results from discriminant function analysis and ANCOVA showed that there were significant differences in external morphology in pumpkinseed captured from the four habitat zones in all three reservoirs. Littoral and pelagic differentiation was stronger than fluvial and lacustrine differentiation in all of the reservoirs, and the most significant variable that differentiated pumpkinseed from the littoral and pelagic habitats was body depth. The illustration of external morphological differentiation in pumpkinseed along both habitat dimensions demonstrates the high degree of morphological plasticity of this introduced species.     

Trophic polymorphism in introduced pumpkinseed (<Emphasis Type="Italic">Lepomis gibbosus</Emphasis>) inhabiting Iberian reservoirs

Introduced pumpkinseed in Iberian reservoirs display marked external morphological differentiation along two simultaneous dimensions of flow and trophic structure. We assessed the degree of internal morphological differentiation using gill rakers, pharyngeal jaws and the levator posterior muscle among pumpkinseeds occupying four different habitats and determined whether prey consumption accounted for any discernible differences in feeding structures among ecomorphs. Results showed significant differentiation by habitat based on pharyngeal muscle and jaw dimensions in all study reservoirs, with pelagic pumpkinseeds having smaller jaws than littoral pumpkinseeds in four of the five reservoirs. Gill rakers, however, differentiated morphs in only three of the five reservoirs, corresponding to differences in zooplankton consumption among pelagic and littoral individuals in those reservoirs. Based on all internal morphological traits, greater divergence was seen along the littoral-pelagic trophic axis in the lacustrine zones of reservoirs compared to the fluvial zone. Overall differences noted in internal morphology are likely the result of phenotypic plasticity; the ability of this species to readily adapt to changing physical environments may explain the success of the pumpkinseed in its introduced range.     

Sympatric Divergence and Performance Trade-Offs of Bluegill Ecomorphs

Phenotypic plasticity in response to environmental cues can create distinct morphological types within populations. This variation in form, and potentially function, may be a factor in initiating population divergence and the formation of new species. Here we show the translation of sympatric, habitat-specific morphological divergence into performance differences in energy economy, maneuverability and steady-state locomotion. Littoral and pelagic bluegill sunfish ecomorphs show differences in performance that appear adaptive within their respective habitats: greater maneuverability in the heavily vegetated littoral; greater steady-state swimming speed and economy in the open-water pelagic. This represents a trade-off in unsteady versus steady swimming performance, likely because morphological features associated with maximizing maneuverability are incompatible with enhancing steady-swimming performance. This may constrain the direction of adaptive change, maintaining the divergence created by phenotypic plasticity. The combination of habitat specific sympatric adaptation and constraints imposed by performance trade-offs may be an important factor underlying the high rate of speciation in freshwater fishes from post-glacial lakes.     

Behavioural Response of Juvenile Bluegill Sunfish to Variation in Predation Risk and Food Level

The behavioural response of juvenile bluegill sunfish (Lepomis macrochirus) to predation risk when selecting between patches of artificial vegetation differing in food and stem density was investigated. Bluegill foraging activity was significantly affected by all three factors. Regardless of patch stem density or risk of predation bluegills preferred patches with the highest prey number. During each trial bluegill foraging activity was clearly divided into a between- and within-patch component. In the presence of a predator bluegills reduced their between-patch foraging activity by an equivalent amount regardless of patch stem density or food level, apparently showing a risk-adjusting behavioural response to predation risk. Within patches, however, foraging activity was affected by both food level and patch stem density. When foraging in a patch offering a refuge from predation, the presence of a predator had no effect on bluegill foraging activity within this patch. However, if foraging in a patch with only limited refuge potential, bluegill foraging activity was reduced significantly in the presence of a predator. Further, this reduction was significantly greater if the patch contained a low versus a high food level, indicating a risk-balancing response to predation with respect to within-patch foraging activity. Both these responses differ from the risk-avoidance response to predation demonstrated by juvenile bluegills when selecting among habitats. Therefore, our results demonstrate the flexibility of juvenile bluegill foraging behaviour.     

Plant-specific expression of antipredator behaviour by larval damselflies

Summary In this study of interactions between larval damselflies (Zygoptera: Coenagrionidae) and pumpkinseed sunfish (Lepomis gibbosus) we focus on the behaviour of the damselfly prey. First we document a prey behaviour in which damselflies use plant stems or leaves to hide from pumpkinseeds. We then test two hypotheses: (1) that damselfly hiding is a specific antipredator behaviour and (2) that hiding occurs more frequently in plant habitats where damselflies experience greater risk of predation. Since plant species growth forms can influence predation risk, our second hypothesis implies that hiding behaviour is conditional upon the type of vegetation providing habitat structure. Conditional expression of antipredator behaviour according to vegetation type may be important in littoral environments, since predator-prey interactions can occur in habitats with a wide range of macrophyte growth forms. The first hypothesis was supported by our findings that damselfly hiding increased in frequency in the presence of pumpkinseeds, that it was related to the frequency of predator approaches, and that its use reduced damselfly predation risk in high risk habitats. The second hypothesis was supported by our results that damselfly hiding rates were greater in the high risk Scirpus habitats than in the lower risk Potamogeton habitats. These results indicate that prey behaviour can influence predator-prey interactions, and that variation in plant growth form can influence prey behaviour, thus contributing to the impact of habitat structure on predator-prey dynamics.     

Squeezing out the last egg—annual fish increase reproductive efforts in response to a predation threat

Both constitutive and inducible antipredator strategies are ubiquitous in nature and serve to maximize fitness under a predation threat. Inducible strategies may be favored over constitutive defenses depending on their relative cost and benefit and temporal variability in predator presence. In African temporary ponds, annual killifish of the genus Nothobranchius are variably exposed to predators, depending on whether larger fish invade their habitat from nearby rivers during floods. Nonetheless, potential plastic responses to predation risk are poorly known. Here, we studied whether Nothobranchius furzeri individuals adjust their life history in response to a predation threat. For this, we monitored key life history traits in response to cues that signal the presence of predatory pumpkinseed sunfish (Lepomis gibbosus). While growth rate, adult body size, age at maturation, and initial fecundity were not affected, peak and total fecundity were higher in the predation risk treatment. This contrasts with known life history strategies of killifish from permanent waters, which tend to reduce reproduction in the presence of predators. Although our results show that N. furzeri individuals are able to detect predators and respond to their presence by modulating their reproductive output, these responses only become evident after a few clutches have been deposited. Overall our findings suggest that, in the presence of a predation risk, it can be beneficial to increase the production of life stages that can persist until the predation risk has faded.     

Effects of perceived predation risk and social environment on the development of three‐spined stickleback (Gasterosteus aculeatus) morphology

Phenotypically plastic changes in response to variation in perceived predation risk are widespread, but little is known about if and how social environment modulates induced responses to predation risk. We investigated the influence of perceived predation risk (i.e. chemical cues from a predator) and social environment (i.e. one, two or 20 individuals reared together) on three‐spined stickleback (Gasterosteus aculeatus) morphology in a factorial common garden experiment. We found that exposure to chemical cues from potential predators did not influence growth or body condition or induce more robust morphological defences (i.e. lateral plate numbers and dorsal spine lengths). However, sticklebacks exposed to predator cues developed longer caudal peduncles and larger eyes as compared with fish from the control treatment. As these responses may improve sticklebacks’ ability to avoid piscine predation, they might be adaptive. Social environment/density also influenced expression of some traits, but these effects were independent of predation‐risk treatment effects. In general, these results suggest that apart from the classic morphological defence structures, which appear mostly constitutive, three‐spined sticklebacks are capable of expressing potentially adaptive morphological responses to chemical cues from potential predators.     

Phenotypic variation and vulnerability to predation in juvenile bluegill sunfish (<Emphasis Type="Italic">Lepomis macrochirus</Emphasis>)

Bluegill sunfish (Lepomis macrochirus) are known to diversify into two forms specialized for foraging on either limnetic or littoral prey. Because juvenile bluegills seek vegetative cover in the presence of largemouth bass (Micropterus salmoides) predators, natural selection should favor the littoral body design at size ranges most vulnerable to predation. Yet within bluegill populations, both limnetic and littoral forms occur where vegetation and predators are present. While adaptive for foraging in different environments, does habitat-linked phenotypic variation also influence predator evasiveness for juvenile bluegills? We evaluate this question by quantifying susceptibility to predation for two groups of morphologically distinct bluegills; a limnetic form characteristic of bluegills inhabiting open water areas (limnetic bluegill) and a littoral form characteristic of bluegills inhabiting dense vegetation (littoral bluegill). In a series of predation trials, we found that bluegill behaviors differed in open water habitat but not in simulated vegetation. In open water habitat, limnetic bluegills formed more dense shoaling aggregations, maintained a larger distance from the predator, and required longer amounts of time to capture than littoral bluegill. When provided with simulated vegetation, largemouth bass spent longer amounts of time pursuing littoral bluegill and captured significantly fewer littoral bluegills than limnetic fish. Hence, morphological and behavioral variation in bluegills was linked to differential susceptibility to predation in open water and vegetated environments. Combined with previous studies, these findings show that morphological and behavioral adaptations enhance both foraging performance and predator evasiveness in different lake habitats.     

Are behavioural responses to predation cues linked across life cycle stages?

1. Prey organisms can perceive cues to predation hazard and adopt low‐risk behaviours to increase survival. Animals with complex life cycles, such as insects, can exhibit such anti‐predatory behaviours in multiple life stages. 2. Cues to predation risk may induce ovipositing females to choose habitats with low predation risk. Cues to predation risk may also induce larvae to adopt facultative behaviours that reduce risk of predation. 3. One hypothesis postulates that anti‐predation behaviours across adult and larval stages may be negatively associated because selection for effective anti‐predator behaviour in one stage leads to reduced selection for avoidance of predators in other stages. An alternative hypothesis suggests that selection by predation favours multi‐component defences, with both avoidance of oviposition and facultative adoption of low‐risk behaviours by larvae. 4. Laboratory and field experiments were used to determine whether defensive responses of adult and larval mosquitoes are positively or negatively associated. The study tested effects of waterborne cues from predatory Toxorhynchites theobaldi on oviposition choices and larval behaviours of three of its common prey: Culex mollis, Limatus durhamii and Aedes albopictus. 5. Culex mollis shows strong anti‐predator responses in both life stages, consistent with the hypothesis of a multi‐component behavioural defence. The other two species showed no detectable responses to waterborne predator cues in either adult or larval stages. Larvae of these unresponsive species were significantly more vulnerable to this predator than was C. mollis. 6. For these mosquitoes, species appear either to have been selected for multi‐component defences against predation or to act in ways that could be called predator‐naïve.     

Sex‐Specific Response of Pardosa milvina (Araneae: Lycosidae) to Experience with a Chemotactile Predation Cue

Predators frequently leave behind chemical information (i.e., semiochemicals such as pheromones or kairomones) that can be detected by their prey and used to avoid areas where predators are likely present. Prey that have interacted indirectly with predators via chemical information thus may gain insight into their risk of being consumed that naïve individuals lack. Pardosa milvina (Araneae: Lycosidae) is a chemosensitive wolf spider that shows adaptive responses to chemotactile cues deposited by the larger wolf spider Tigrosa helluo. We raised offspring from P. milvina to examine the effect of experience with a predation cue on activity, foraging, and antipredator behavior. Spiders differed in activity and foraging behavior across ontogeny and between sexes, but there was no effect of experience with a predation cue. However, a sex‐specific effect of experience was found in antipredator behavior. Male spiders, but not females, used experience with a predator cue to increase their survival in the presence of a live predator. Specifically, naïve males were attacked sooner than experienced males, indicating that prior exposure to predator cues can modify Pardosa antipredator behavior. Intersexual differences in how spiders respond to experience with a predation cue likely reflect the risk of predation faced by males and females in nature.     

Survival Benefit to Antipredator Behavior in the Amphipod Gammarus minus (Crustacea: Amphipoda) in Response to Injury-released Chemical Cues from Conspecifics and Heterospecifics

A wide range of aquatic taxa use environmental chemical cues for the assessment of predation risk. We examined whether Gammarus minus (Crustacea: Amphipoda) exhibit antipredator behavior in response to injury-released chemicals from conspecifics or heterospecifics (Crustacea: Isopoda). We then examined whether behavioral responses to these cues conferred survival benefits to the amphipods. In the first part of this study, we tested the behavioral response of G. minus to the following treatments: 1. water containing injury-released cues of conspecifics; 2. water containing injury-released cues of a sympatric isopod crustacean, Lirceus fontinalis; or 3. water containing no cues (control). Relative to the control, Gammarus responded to the conspecific cue by moving to the substratum and decreasing activity. In contrast, Gammarus responded to the heterospecific cue by moving up into the water column and increasing activity. In the second part of this study, we tested if the behavioral response to these cues confers a survival benefit to Gammarus when exposed to a predator. A green sunfish ( Lepomis cyanellus ) was retained behind a partition in the test tanks. Two minutes after the introduction of the chemical cues in the first test, the barrier was lifted and predation events recorded. Relative to the control, the time to the first attack increased for Gammarus exposed to conspecific cues and decreased for those exposed to heterospecific cues. These data indicate that Gammarus distinguish between chemical cues from conspecific and heterospecific crustaceans, and that the antipredator response to conspecific cues confers a fitness benefit (i.e. increased survival due to increased time to the first attack).     

Genetic variation and phenotypic plasticity in a trophically polymorphic population of pumpkinseed sunfish (Lepomis gibbosus)

Summary Adaptive variation can exist at a variety of scales in biological systems, including among species, among local populations of a single species and among individuals within a single population. Trophic or resource polymorphisms in fishes are a good example of the lowest level of this hierarchy. In lakes without bluegill sunfish (Lepomis macrochirus), pumpkinseed sunfish (Lepomis gibbosus) can be trophically polymorphic, including a planktivorous limnetic form found in the pelagic habitat, in addition to the usual benthic form found in the littoral zone. In this paper we examine the degree to which morphological differences between the two forms are caused by genetic differences versus phenotypic plasticity. Adults from pelagic and littoral sites in Paradox Lake, NY, were bred separately and their progeny were raised in cages both in the open water and shallow water habitats of an artificial pond. The experimental design permitted two tests of genetic differences between the breeding stocks (in open and shallow water cages, respectively) and two tests of phenotypic plasticity (in the limnetic and benthic offspring, respectively). Limnetic progeny were more fusiform than benthic progeny raised in the same habitat. In addition, progeny of both stocks displayed limnetic-type characteristics when raised in the open water and benthic-type characteristics in the shallow water. Thus, genetic differences and phenotypic plasticity both contributed to the trophic polymorphism. Phenotypic plasticity and genetic differentiation accounted for 53 and 14%, respectively, of the variation in morphology. This study addresses the nature of subtle phenotypic differences among individuals from a single population that is embedded within a complex community, a condition that is likely to be the norm for most natural populations, as opposed to very large differences that have evolved in relatively few populations that reside in species-poor environments.     

Nature of Predation Risk Cues in Container Systems: Mosquito Responses to Solid Residues From Predation

In aquatic systems, prey animals associate predation risk with cues that originate either from the predator or from injured conspecifics. Sources and benefits of these cues have received considerable attention in river, lake, and pond ecosystems but are less well understood in small container ecosystems that can hold less than a liter of water. Mosquitoes Aedes triseriatus (Say) and Aedes albopictus (Skuse) encounter predatory Corethrella appendiculata (Grabham) and Toxorhynchites rutilus (Coquillett) in small containers and show antipredatory behavioral responses. We investigated the sources of the predation cues to which these prey larvae respond. We tested whether Ae. albopictus larvae show behavioral responses to cues emanating from the predator or from damage to prey caused by the act of predation. We also tested whether Ae. triseriatus respond to cues present in fluid or solid residues from predator activity. Ae. albopictus showed behavioral modifications only in response to waterborne cues from a feeding predator and not to cues from a starving predator, indicating that Ae. albopictus respond to cues created by the act of predation, which could include substances derived from damaged prey or substances in predator feces. Ae. triseriatus showed behavioral responses to solid residues from predation but not to fluid without those solids, indicating that the cues to which they respond originate in predator feces or uneaten prey body parts. Our results suggest that cues in this system may be primarily chemicals that are detected upon contact with solid residues that are products of the feeding processes of these predators.     

Predator Diversity Effects in an Exotic Freshwater Food Web

Cascading trophic interactions are often defined as the indirect effects of a predator on primary producers through the effect of the predator on herbivores. These effects can be both direct through removal of herbivores [density-mediated indirect interactions (DMIIs)] or indirect through changes in the behavior of the herbivores [trait-mediated indirect interactions (TMIIs)]. How the relative importance of these two indirect interactions varies with predator diversity remains poorly understood. We tested the effect of predator diversity on both TMIIs and DMIIs on phytoplankton using two competitive invasive dreissenid mussel species (zebra mussel and quagga mussel) as the herbivores and combinations of one, two or all three species of the predators pumpkinseed sunfish, round goby, and rusty crayfish. Predators had either direct access to mussels and induced both TMII and DMII, or no direct access and induced only TMII through the presence of risk cues. In both sets of treatments, the predators induced a trophic cascade which resulted in more phytoplankton remaining with predators present than with only mussels present. The trophic cascade was weaker in three-predator and two-predator treatments than in one-predator treatments when predators had direct access to dreissenids (DMIIs and TMIIs). Crayfish had higher cascading effects on phytoplankton than both pumpkinseed and round goby. Increased predator diversity decreased the strength of DMIIs but had no effect on the strength of TMIIs. The strength of TMIIs was higher with zebra than quagga mussels. Our study suggests that inter-specific interference among predators in multi-species treatments weakens the consumptive cascading effects of predation on lower trophic levels whereas the importance of predator diversity on trait mediated effects depends on predator identity.