Adaptive evolution to novel predators facilitates the evolution of damselfly species range shifts

Most species have evolved adaptations to reduce the chances of predation. In many cases, adaptations to coexist with one predator generate tradeoffs in the ability to live with other predators. Consequently, the ability to live with one predator may limit the geographic distributions of species, such that adaptive evolution to coexist with novel predators may facilitate range shifts. In a case study with Enallagma damselflies, we used a comparative phylogenetic approach to test the hypothesis that adaptive evolution to live with a novel predator facilitates range size shifts. Our results suggest that the evolution of Enallagma shifting from living in ancestral lakes with fish as top predators, to living in lakes with dragonflies as predators, may have facilitated an increase in their range sizes. This increased range size likely arose because lakes with dragonflies were widespread, but unavailable as a habitat throughout much of the evolutionary history of Enallagma because they were historically maladapted to coexist with dragonfly predators. Additionally, the traits that have evolved as defenses against dragonflies also likely enhanced damselfly dispersal abilities. While many factors underlie the evolutionary history of species ranges, these results suggest a role for the evolution of predator‐prey interactions.     

PREDATOR‐DRIVEN TRAIT DIVERSIFICATION IN A DRAGONFLY GENUS: COVARIATION IN BEHAVIORAL AND MORPHOLOGICAL ANTIPREDATOR DEFENSE

Proof for predation as an agent shaping evolutionary trait diversification is accumulating, however, our understanding how multiple antipredator traits covary due to phenotypic differentiation is still scarce. Species of the dragonfly genus Leucorrhinia underwent shifts from lakes with fish as top predators to fishless lakes with large dragonfly predators. This move to fishless lakes was accompanied by a partial loss and reduction of larval spines. Here, we show that Leucorrhinia also reduced burst swimming speed and its associated energy fuelling machinery, arginine kinase activity, when invading fishless lakes. This results in patterns of positive phylogenetic trait covariation between behavioral and morphological antipredator defense (trait cospecialization) and between behavioral antipredator defense and physiological machinery (trait codependence). Across species patterns of trait covariation between spine status, burst swimming speed and arginine kinase activity also matched findings within the phenotypically plastic L. dubia. Our results highlight the importance of predation as a factor affecting patterns of multiple trait covariation during phenotypic diversification.     

MORPHOLOGICAL EVOLUTION MEDIATED BY BEHAVIOR IN THE DAMSELFLIES OF TWO COMMUNITIES

Behavior can play a mediating role in determining the selective pressures that influence the evolution of morphological structures. To examine this, I quantified patterns of morphological variation among larvae of Enallagma damselfly species (Odonata: Coenagrionidae) that use different behaviors to avoid the major predators found in each of two communities, lakes with and without fish. Specifically, I quantified the sizes and shapes of the abdomens and caudal lamellae (used for swimming) and legs for three species from fishless lakes and six species from lakes with fish. A preliminary cladistic analysis indicates that species within each lake type are not members of a single clade, which supports the conclusions of previous odonate taxonomists. Previous studies have shown that species in fishless lakes are very active, running and swimming frequently and at high rates of speed in the absence of predators, and they avoid their primary predators, large dragonflies, by swimming. These species have the widest abdomens, the largest caudal lamellae relative to overall body size, and the longest legs of the species studied, which should make them powerful swimmers and runners. Furthermore, species in fishless lakes are morphologically very similar to one another and differ greatly from fish-lake species, although each is more closely related to species in fish lakes. In contrast, species from lakes with fish move very slowly and infrequently in the absence of predators and do not attempt to evade attacking predators. However, despite their behavioral similarity, large interspecific variation in morphology exists among the fish-lake species, and the only morphological patterns were differences associated with membership in the two primary clades identified in the cladistic analysis. A modification of Felsenstein's (1985) method of evolutionary contrasts which allows character change to be isolated along single branches is introduced and is used to reconstruct the evolutionary histories of these characters. This analysis suggests that large increases in caudal lamella size, abdominal segment lengths and widths, and leg length accompany speciation events associated with habitat shifts from fish-lakes to fishless lakes. Following habitat shifts selection pressures exerted by dragonfly predation apparently favored swimming as an escape tactic, which mediated selection pressures onto morphologies used in swimming to increase swimming performance; morphological patterns in extant species reflect this adaptation to a new environment. Mechanisms by which behaviorally mediated selection could have accelerated evolutionary dynamics following founder events are discussed.     

PREDISPOSED TO ADAPT? CLADE‐LEVEL DIFFERENCES IN CHARACTERS AFFECTING SWIMMING PERFORMANCE IN DAMSELFLIES

Previous studies have shown that two or three lineages of Enallagma damselflies, which historically co-existed with fish, recently invaded and adapted to living with large dragonfly predators in fishless waters. In adapting to live with these new predators, lineages shifted behaviorally to using swimming as an evasive tactic against attacking predators, evolved morphological features that made them faster swimmers, and evolved biochemical features to increase refueling strenuous activities like swimming. However, these habitat shifts have occurred in only one of the two primary clades within the genus in North America. Here, I show that clade-level differences exist among species in the ancestral, fish-lake habitat that should make habitat shifts easier to accomplish in the clade in which they have occurred. Specifically, fish-lake species in the clade in which habitat shifts occurred have much higher propensities to swim in the laboratory, swim faster when they do swim, and have higher mass-specific activities for arginine kinase than do species in the other primary clade, in which no extant species are found in fishless waters. These results are discussed in the context of the dynamics of founder events and the potential implications for community structure.     

MEASURING PHENOTYPIC SELECTION ON AN ADAPTATION: LAMELLAE OF DAMSELFLIES EXPERIENCING DRAGONFLY PREDATION

Previous studies suggest that the evolution of increased caudal lamellae size to increase swimming speed was an adaptation of Enallagma damselflies for coexisting with large, predatory dragonflies in fishless lakes. To test whether dragonfly predation still exerts selection pressures for increased lamellae size, I performed a field experiment in which I manipulated the abilities of dragonfly larvae to inflict mortality on Enallagma boreale larvae and compared differences in lamellae size and shape between treatments. In cages where dragonflies were free to forage on damselflies, surviving E. boreale larvae had lamellae that were larger in lateral surface area, and that were wider relative to their length, as compared with larvae recovered from treatments in which dragonflies were not permitted to forage on damselflies. Selection differentials of about 0.25 phenotypic standard deviation units were measured for both of these characters. These results indicate that dragonfly predation still exerts significant selection pressures on damselfly antipredator adaptations. The results of this study are discussed in the context of studies of adaptation.     

Evolution of prey behavior in response to changes in predation regime: damselflies in fish and dragonfly lakes

In a large behavioral experiment we reconstructed the evolution of behavioral responses to predators to explore how interactions with predators have shaped the evolution of their prey's behavior. All Enallagma damselfly species reduced both movement and feeding in the presence of coexisting predators. Some Enallagma species inhabit water bodies with both fish and dragonflies, and these species responded to the presence of both predators, whereas other Enallagma species inhabit water bodies that have only large dragonflies as predators, and these species only responded to the presence of dragonflies. Lineages that shifted to live with large dragonflies showed no evolution in behaviors expressed in the presence of dragonflies, but they evolved greater movement in the absence of predators and greater movement and feeding in the presence of fish. These results suggest that Enallagma species have evolutionarily lost the ability to recognize fish as a predator. Because species coexisting with only dragonfly predators have also evolved the ability to escape attacking dragonfly predators by swimming, the decreased predation risk associated with foraging appears to have shifted the balance of the foraging/predation risk trade-off to allow increased activity in the absence of mortality threats to evolve in these lineages. Our results suggest that evolution in response to changes in predation regime may have greater consequences for characters expressed in the absence of mortality threats because of how the balance between the conflicting demands of growth and predation risk are altered.     

Changing the habitat: the evolution of intercorrelated traits to escape from predators

Burst escape speed is an effective and widely used behaviour for evading predators, with burst escape speed relying on several different morphological features. However, we know little about how behavioural and underlying morphological attributes change in concert as a response to changes in selective predation regime. We studied intercorrelated trait differentiation of body shape and burst‐swim‐mediating morphology in response to a habitat shift‐related reduction in burst escape speed using larvae of the dragonfly genus Leucorrhinia. Species in this genus underwent a well‐known habitat shift from predatory fish lakes (fish lakes) to predatory fish‐free lakes dominated by large predatory dragonflies (dragonfly lakes) accompanied by relaxed selection on escape burst speed. Results revealed that species from fish lakes that possess faster burst speed have evolved a suite of functionally intercorrelated traits, expressing a wider abdomen, a higher abdominal muscles mass and a larger branchial chamber compared with species from dragonfly lakes. In contrast, populations within species did not show significant differences in muscle mass and branchial chamber size between lake types in three of the species. High multicollinearity among variables suggests that traits have evolved in concert rather than independently when Leucorrhinia shifted from fish lakes to dragonfly lakes. Thus, relaxed selection on burst escape speed in dragonfly‐lake species resulted in a correlated reduction of abdominal muscles and a smaller branchial chamber, likely to save production and/or maintenance costs. Our results highlight the importance of studying integrated behavioural and morphological traits to fully understand the evolution of complex phenotypes.     

Relaxed predation results in reduced phenotypic integration in a suite of dragonflies

Although changes in magnitude of single traits responding to selective agents have been studied intensively, little is known about selection shaping networks of traits and their patterns of covariation. However, this is central for our understanding of phenotypic evolution as traits are embedded in a multivariate environment with selection affecting a multitude of traits simultaneously rather than individually. Here, we investigate inter‐ and intraspecific patterns of trait integration (trait correlations) in the larval abdomen of dragonflies as a response to a change in predator selection. Species of the dragonfly genus Leucorrhinia underwent a larval habitat shift from predatory fish to predatory dragonfly‐dominated lakes with an associated relaxation in selection pressure from fish predation. Our results indicate that the habitat‐shift‐induced relaxed selection pressure caused phenotypic integration of abdominal traits to be reduced. Intraspecific findings matched patterns comparing species from both habitats with higher abdominal integration in response to predatory fish. This higher integration is probably a result of faster burst swimming speed. The abdomen holds the necessary morphological machinery to successfully evade predatory fish via burst swimming. Hence, abdominal traits have to function in a tight coordinated manner, as maladaptive variation and consequently nonoptimal burst swimming would cause increased mortality. In predatory dragonfly‐dominated lakes, no such strong link between burst swimming and mortality is present. Our findings highlight the importance of studying multivariate trait relationships as a response to selection for understanding patterns of phenotypic diversification.     

Shift in predation regime mediates diversification of foraging behaviour in a dragonfly genus

1. Behavioural adaptations to avoid and evade predators are common. Many studies have investigated population divergence in response to changes in predation regime within species, but studies exploring interspecific patterns are scant. Studies on interspecific divergence can infer common outcomes from evolutionary processes and highlight the role of environmental constraints in shaping species traits. 2. Species of the dragonfly genus Leucorrhinia underwent well‐studied shifts from habitats being dominated by predatory fish (fish lakes) to habitat being dominated by predatory invertebrates (dragonfly lakes). This change in top predators resulted in a set of adaptive trait modifications in response to the different hunting styles of both predator types: whereas predatory fish actively search and pursue prey, invertebrate predator follow a sit‐and‐wait strategy, not pursuing prey. 3. Here it is shown that the habitat shift‐related change in selection regime on larval Leucorrhinia caused species in dragonfly lakes to evolve increased larval foraging and activity, and results suggest that they lost the ability to recognise predatory fish. 4. The results of the present study highlight the impact of predators on behavioural trait diversification with habitat‐specific predation regimes selecting for distinct behavioural expression.     

Flexible antipredator behavior in a dragonfly species that coexists with different predator types

Two of the main predators of dragonfly larvae, insectivorous fish in communities with fish and large dragonfly species in communities without fish, differ markedly in their mode of predation. In general, dragonfly species coexist successfully with one predator or the other, but larvae of the dragonfly Pachydiplax longipennis can coexist successfully with both. I examined the behavioral response of these larvae to a simulated predator attack to determine whether their response (1) differs between the two communities, and (2) is sensitive to waterborne cues about the type of predator present. I compared larvae from two different communities: fish ponds where insectivorous fish were the top predators, and fish-free ponds where large dragonflies were the top predators. Larvae from fish-free ponds actively moved away from an attack significantly more than did larvae from fish ponds, provided each was attacked in its native pond water. Larvae collected from a fish-free pond but then attacked in fish water moved less than did controls (larvae attacked in fish-free water). Likewise, larvae collected from a fish pond but attacked in fish-free water moved more than did controls (larvae attacked in fish water). Larvae attacked first in water from their native pond and then in water from the contrasting pond changed their response in the expected direction. These results indicate that escape behavior in P. longipennis differs between communities with different predator types and is sensitive to waterborne cues in a manner consistent with the mode of predation employed by each predator.     

Predator-induced morphological changes in an amphibian: predation by dragonflies affects tadpole shape and color

Predator-induced defenses are well studied in plants and invertebrate animals, but have only recently been recognized in vertebrates. Gray treefrog (Hylachrysoscelis) tadpoles reared with predatory dragonfly (Aeshnaumbrosa) larvae differ in shape and color from tadpoles reared in the absence of dragonflies. By exposing tadpoles to tail damage and the non-lethal presence of starved and fed dragonflies, we determined that these phenotypic differences are induced by non-contact cues present when dragonflies prey on Hyla. The induced changes in shape are in the direction that tends to increase swimming speed; thus, the induced morphology may help tadpoles evade predators. Altering morphology in response to predators is likely to influence interactions with other species in the community as well. Received: 17 April 1996 / Accepted: 18 September 1996     

Specific induced responses to different predator species in anuran larvae

Models of defence against multiple enemies predict that specialized responses to each enemy should evolve only under restrictive conditions. Nevertheless, tadpoles of Rana temporaria can differentiate among several predator species. Small tadpoles used a refuge when Notonecta backswimmers were in the pond, but showed a weaker hiding response to adult Triturus alpestris newts and no response to aeshnid dragonfly larvae (Aeshna and Anax). All predators caused a decline in feeding and swimming activity. Large tadpoles reserved the strongest behavioural response for dragonflies, while Triturus caused no response. The shift during development suggests that tadpoles distinguished among predators, rather than exhibiting a graded dosage response to a single cue associated with predation. Information on habitat distributions of predators suggests that they are regularly encountered, which would facilitate evolution of adaptive behavioural responses. Morphological responses to all predators were similar, perhaps because similar morphologies defend against all four predators. The evolutionary maintenance of specialized responses to multiple predators may be possible because adaptive responses do not conflict and the predators themselves do not interact strongly.     

Invertebrate predation predicts variation in an autotomy-related trait in larval damselfly

Autotomy, the discarding of a prey appendage grasped by a predator, may evolve when the benefits of immediate survival outweigh the costs of appendage loss. In larval damselflies, joints connecting lamellae to the abdomen vary in size and shape within and among taxa suggesting that they may evolve under selection by invertebrate predators, such as dragonfly larvae. Assuming that joint width is proportional to the force required for autotomy, we tested if invertebrate predation favours smaller lamellar joints for autotomy or larger joints for structural support of lamellae for swimming. We compared the maximum joint widths of larval Lestes and Enallagma among ponds that varied in risk of invertebrate predation. Relative predation risk estimated as the frequency of regenerated lamellae within ponds was weakly and positively related to the relative abundance of larval dragonflies. The allometry of lamellar joint size decreased with increasing risk of invertebrate predation across ponds after controlling for variation in body size in Lestes congener but not in Enallagma species. Both species of Lestes had larger joint sizes than the five species of Enallagma, suggesting that the ancestral divergence of lamellar joints between these genera may influence contemporary responses to invertebrate predation.     

Parallel evolution in ecological and reproductive traits to produce cryptic damselfly species across the holarctic

The damselfly genus Enallagma originated in the Nearctic, and two Nearctic lineages recently underwent radiations partly associated with multiple independent habitat shifts from lakes dominated by fish predators into lakes dominated by dragonfly predators. A previous molecular study of four Palearctic morphospecies and all representative Nearctic species identified the presence of two cryptic species sets, with each set having Palearctic and Nearctic representatives. However, the cryptic species within each set are not sibling species. Here, we present quantitative data on ecologically important larval morphologies and behaviors involved in predator avoidance and on adult male morphological structures involved in mate recognition to quantify the phenotypic relationships among these cryptic species sets. For the adult stage, our data indicate strong parallel evolution of the structures involved in specific mate recognition-the male cerci. For the larval stage, morphometric analyses show that the Palearctic species evolved a nearly identical morphology to the sibling-clade members in the Nearctic that live in waters where dragonflies are the top predators. This implicates the importance of dragonfly predation in the history of the Palearctic clade. Behavioral analyses suggest population differentiation in response to the actual predator environment in the Palearctic clade, consistent with the species differentiation seen in the Nearctic. Our results suggest parallel evolution of adult traits that influence specific mate choice and larval traits that influence ecological performance underlie the striking similarity of Enallagma species across continents. This concurrent parallel evolution in both stages of a complex life cycle, especially when both stages do not share the same selective environment, may be a very unusual mechanism generating cryptic species.     

Prey detection of aquatic predators: Assessing the identity of chemical cues eliciting prey behavioral plasticity

Chemical cues transmitted through the environment are thought to underlie many prey responses to predation risk, but despite the known ecological and evolutionary significance of such cues, their basic composition are poorly understood. Using anuran tadpoles (prey) and dragonfly larvae (predators), we identified chemical cues associated with predation risk via solid phase extraction and mass spectrometry of the extracts. We found that dragonfly larvae predators consistently produced a negative ion, m/z 501.3, when they fed on bullfrog (Rana catesbeiana) and mink frog (Rana septentrionalis) tadpoles, but this ion was absent when dragonflies were fasted or fed invertebrate prey. When tadpole behavioral responses to dragonfly chemical cues were examined, tadpoles reduced their activity, particularly in response to dragonflies feeding on tadpoles. Furthermore, a negative correlation was noted between the level of tadpole activity and the concentration of the m/z 501.3 compound in dragonfly feeding trials, indicating that this ion was possibly responsible for tadpole anti-predator behavior.     

Dragonfly predators influence biomass and density of pond snails

Studies in lakes show that fish and crayfish predators play an important role in determining the abundance of freshwater snails. In contrast, there are few studies of snails and their predators in shallow ponds and marshes. Ponds often lack fish and crayfish but have abundant insect populations. Here we present the results of field surveys, laboratory foraging trials, and an outdoor mesocosm experiment, testing the hypothesis that insects are important predators of pulmonate snails. In laboratory foraging trials, conducted with ten species of insects, most insect taxa consumed snails, and larval dragonflies were especially effective predators. The field surveys showed that dragonflies constitute the majority of the insect biomass in fishless ponds. More focused foraging trials evaluated the ability of the dragonflies Anax junius and Pantala hymenaea to prey upon different sizes and species of pulmonate snails (Helisoma trivolvis, Physa acuta, and Stagnicola elodes). Anax junius consumed all three species up to the maximum size tested. Pantala hymenaea consumed snails with a shell height of 3 mm and smaller, but did not kill larger snails. P. acuta were more vulnerable to predators than were H. trivolvis or S. elodes. In the mesocosm experiment, conducted with predator treatments of A. junius, P. hymenaea, and the hemipteran Belostoma flumineum, insect predators had a pronounced negative effect on snail biomass and density. A. junius and B. flumineum reduced biomass and density to a similar degree, and both reduced biomass more than did P. hymenaea. Predators did not have a strong effect on species composition. A model suggested that A. junius and P. hymenaea have the largest effects on snail biomass in the field. Given that both pulmonate snails and dragonfly nymphs are widespread and abundant in marshes and ponds, snail assemblages in these water bodies are likely regulated in large part by odonate predation.     

Estimation of effective concentrations of ATP-regenerating enzymes in cilia of Paramecium caudatum

The phosphoarginine shuttle system effectively regenerates ATP in the cilia of Paramecium caudatum. To estimate the effective concentration of ATP‐regenerating enzymes, we attempted to reconstitute certain ATP‐regenerating systems within the cilia of intact cortical sheets using exogenous enzymes and high‐energy substances. The addition of phosphoenolpyruvate, which is one of the substrates in glycolysis, did not increase the ciliary beat frequency, whereas phosphocreatine together with exogenous creatine kinase, effectively increased the ciliary beat frequency. In the presence of 0.6 mg/ml creatine kinase and 0.4 mM phosphocreatine, the ciliary beat frequency was comparable to that produced by the addition of phosphoarginine. This result indicates that the reconstituted phosphocreatine shuttle system can work as an artificial ATP‐regenerating system for ciliary movements. The effective concentration of creatine kinase in the reconstituted phosphocreatine shuttle system was estimated to be about 7.4 μM based on the molecular mass of creatine kinase (MW 81,000). Therefore, the effective concentration of arginine kinase in the cilia of live Paramecium is approximately 10 μM. This estimated concentration of intraciliary arginine kinase is sufficient to maintain a high ATP concentration throughout the cilia of P. caudatum.     

Alternative growth and energy storage responses to mortality threats in damselflies

The role of physiology in mediating the growth/predation risk trade‐off has been largely ignored. We examined effects of predation risk on relationships between growth and storage molecules in Enallagma aspersum and Ischnura verticalis damselfly larvae that differ in this trade‐off. In laboratory and field experiments, both species had similar growth and mortality rates and similar concentrations of storage molecules in the absence of mortality threats. However, in the presence of dragonfly predators Ischnura larvae had higher mortality rates and grew faster than Enallagma larvae. Consistent with the difference in growth rate, Enallagma's total protein concentrations decreased under predation risk while those of Ischnura did not. Glucose and glycogen concentrations were not affected, while triglyceride concentrations were lower under predation risk in Enallagma but not in Ischnura. Species differences at the physiological level to the presence of mortality threats may be crucial to understanding patterns in metamorphic and post‐metamorphic traits.     

Past and Present Risk: Exposure to Predator Chemical Cues Before and after Metamorphosis Influences Juvenile Wood Frog Behavior

Animals are exposed to different predators over their lifespan. This raises the question of whether exposure to predation risk in an early life stage affects the response to predators in subsequent life stages. In this study, we used wood frogs (Rana sylvatica) to test whether exposure to cues indicating predation risk from dragonfly larvae during the wood frog larval stage affected post‐metamorphic activity level and avoidance of garter snake chemical cues. Dragonfly larvae prey upon wood frogs only during the larval stage, whereas garter snakes prey upon wood frogs during both the larval stage and the post‐metamorphic stage. Exposure to predation risk from dragonflies during the larval stage caused post‐metamorphic wood frog juveniles to have greater terrestrial activity than juvenile wood frogs that were not exposed to larval‐stage predation risk from dragonflies. However, exposure to predation risk as larvae did not affect juvenile wood frog responses to chemical cues from garter snakes. Wood frogs exposed as larvae to predation risk from dragonfly larvae avoided garter snake chemical cues to the same extent as wood frog larvae not exposed to predation risk from dragonfly larvae. Our results demonstrate that while some general behaviors exhibit carry‐over effects from earlier life stages, behavioral responses to predators may remain independent of conditions experienced in earlier life stages.     

Prey pursuit and interception in dragonflies

Perching dragonflies (Libellulidae; Odonata) are sit-and-wait predators, which take off and pursue small flying insects. To investigate their prey pursuit strategy, we videotaped 36 prey-capture flights of male dragonflies, Erythemis simplicicollis and Leucorrhinia intacta, for frame-by-frame analysis. We found that dragonflies fly directly toward the point of prey interception by steering to minimize the movement of the prey's image on the retina. This behavior could be guided by target-selective descending interneurons which show directionally selective visual responses to small-object movement. We investigated how dragonflies discriminate distance of potential prey. We found a peak in angular velocity of the prey shortly before take-off which might cue the dragonfly to nearby flying targets. Parallax information from head movements was not required for successful prey pursuit. Accepted: 11 November 1999