INFLUENCE OF THE INDIRECT EFFECTS OF GUPPIES ON LIFE‐HISTORY EVOLUTION IN RIVULUS HARTII

Early theories of life‐history evolution predict that increased predation on young/small individuals selects for delayed maturation and decreased reproductive effort, but such theory only considers changes in mortality. Predators reduce prey abundance and increase food to survivors. Theory that incorporates such indirect effects yields different predictions. Trinidadian killifish, Rivulus hartii, inhabit communities with and without guppies. Guppies prey on young Rivulus and Rivulus densities decline and growth rates increase when guppies are present. Prior work showed that Rivulus phenotypes from communities with guppies matured earlier and had higher fecundity, consistent with theories that incorporate indirect effects. Here we examined the genetic basis of these differences by rearing 2nd generation, laboratory‐born Rivulus from sites with and without guppies under two food levels that match natural differences in growth. Many locality × food interactions were significant, often reversing the relationship between communities. Such interactions imply that there are fitness trade‐offs associated with adaptation to high or low resource environments. On high food, Rivulus from localities with guppies matured earlier, produced many small eggs, and exhibited increased reproductive investment; these differences reversed on low food. Our results suggest that indirect effects mold Rivulus evolution and thereby highlight connections between community processes and evolutionary change.     

A minimum stochastic model evaluating the interplay between population density and drift for species coexistence

Despite the general acknowledgment of the role of niche and stochastic process in community dynamics, the role of species relative abundances according to both perspectives may have different effects regarding coexistence patterns. In this study, we explore a minimum probabilistic stochastic model to determine the relationship of populations relative and total abundances with species chances to outcompete each other and their persistence in time (i.e., unstable coexistence). Our model is focused on the effects drift (i.e., random sampling of recruitment) under different scenarios of selection (i.e., fitness differences between species). Our results show that taking into account the stochasticity in demographic properties and conservation of individuals in closed communities (zero-sum assumption), initial population abundance can strongly influence species chances to outcompete each other, despite fitness inequalities between populations, and also, influence the period of coexistence of these species in a particular time interval. Systems carrying capacity can have an important role in species coexistence by exacerbating fitness inequalities and affecting the size of the period of coexistence. Overall, the simple stochastic formulation used in this study demonstrated that populations initial abundances could act as an equalizing mechanism, reducing fitness inequalities, which can favor species coexistence and even make less fitted species to be more likely to outcompete better-fitted species, and thus to dominate ecological communities in the absence of niche mechanisms. Although our model is restricted to a pair of interacting species, and overall conclusions are already predicted by the Neutral Theory of Biodiversity, our main objective was to derive a model that can explicitly show the functional relationship between population densities and community mono-dominance odds. Overall, our study provides a straightforward understanding of how a stochastic process (i.e., drift) may affect the expected outcome based on species selection (i.e., fitness inequalities among species) and the resulting outcome regarding unstable coexistence among species.     

Trait adaptation promotes species coexistence in diverse predator and prey communities

Species can adjust their traits in response to selection which may strongly influence species coexistence. Nevertheless, current theory mainly assumes distinct and time‐invariant trait values. We examined the combined effects of the range and the speed of trait adaptation on species coexistence using an innovative multispecies predator–prey model. It allows for temporal trait changes of all predator and prey species and thus simultaneous coadaptation within and among trophic levels. We show that very small or slow trait adaptation did not facilitate coexistence because the stabilizing niche differences were not sufficient to offset the fitness differences. In contrast, sufficiently large and fast trait adaptation jointly promoted stable or neutrally stable species coexistence. Continuous trait adjustments in response to selection enabled a temporally variable convergence and divergence of species traits; that is, species became temporally more similar (neutral theory) or dissimilar (niche theory) depending on the selection pressure, resulting over time in a balance between niche differences stabilizing coexistence and fitness differences promoting competitive exclusion. Furthermore, coadaptation allowed prey and predator species to cluster into different functional groups. This equalized the fitness of similar species while maintaining sufficient niche differences among functionally different species delaying or preventing competitive exclusion. In contrast to previous studies, the emergent feedback between biomass and trait dynamics enabled supersaturated coexistence for a broad range of potential trait adaptation and parameters. We conclude that accounting for trait adaptation may explain stable and supersaturated species coexistence for a broad range of environmental conditions in natural systems when the absence of such adaptive changes would preclude it. Small trait changes, coincident with those that may occur within many natural populations, greatly enlarged the number of coexisting species.     

Nonlinear frequency‐dependent selection promotes long‐term coexistence between bacteria species

Negative frequency‐dependent selection (NFDS) is an important mechanism for species coexistence and for the maintenance of genetic polymorphism. Long‐term coexistence nevertheless requires NFDS interactions to be resilient to further evolution of the interacting species or genotypes. For closely related genotypes, NFDS interactions have been shown to be preserved through successive rounds of evolution in coexisting lineages. On the contrary, the evolution of NFDS interactions between distantly related species has received less attention. Here, we tracked the co‐evolution of Escherichia coli and Citrobacter freundii that initially differ in their ecological characteristics. We showed that these two bacterial species engaged in an NFDS interaction particularly resilient to further evolution: despite a very strong asymmetric rate of adaptation, their coexistence was maintained owing to an NFDS pattern where fitness increases steeply as the frequency decreases towards zero. Using a model, we showed how and why such NFDS pattern can emerge. These findings provide a robust explanation for the long‐term maintenance of species at very low frequencies.     

Parting ways: parasite release in nature leads to sex‐specific evolution of defence

We evaluated the extent to which males and females evolve along similar or different trajectories in response to the same environmental shift. Specifically, we used replicate experimental introductions in nature to consider how release from a key parasite (Gyrodactylus) generates similar or different defence evolution in male vs. female guppies (Poecilia reticulata). After 4–8 generations of evolution, guppies were collected from the ancestral (parasite still present) and derived (parasite now absent) populations and bred for two generations in the laboratory to control for nongenetic effects. These F2 guppies were then individually infected with Gyrodactylus, and infection dynamics were monitored on each fish. We found that parasite release in nature led to sex‐specific evolutionary responses: males did not show much evolution of resistance, whereas females showed the evolution of increased resistance. Given that male guppies in the ancestral population had greater resistance to Gyrodactylus than did females, evolution in the derived populations led to reduction of sexual dimorphism in resistance. We argue that previous selection for high resistance in males constrained (relative to females) further evolution of the trait. We advocate more experiments considering sex‐specific evolutionary responses to environmental change.     

Toward an ecological synthesis: a case for habitat selection

Habitat selection, and its associated density and frequency-dependent evolution, has a profound influence on such vital phenomena as population regulation, species interactions, the assembly of ecological communities, and the origin and maintenance of biodiversity. Different strategies of habitat selection, and their importance in ecology and evolution, can often be revealed simply by plots of density in adjacent habitats. For individual species, the strategies are closely intertwined with mechanisms of population regulation, and with the persistence of populations through time. For interacting species, strategies of habitat selection are not only responsible for species coexistence, but provide one of the most convenient mechanisms for measuring competition, and the various community structures caused by competitive interactions. Other kinds of interactions, such as those between predators and prey, demonstrate that an understanding of the coevolution of habitat-selection strategies among strongly interacting species is essential to properly interpret their spatial and temporal dynamics. At the evolutionary scale, the frequency dependence associated with habitat selection may often allow populations to diverge and diversify into separate species. Habitat selection thereby demonstrates how we can map microevolutionary strategies in behavior onto their population and community consequences, and from there, onto macroevolutionary patterns of speciation and adaptive radiation. We can anticipate that future studies of habitat selection will not only help us complete those maps, but that they will also continue to enrich the panoply of ideas that shape evolutionary ecology.     

The eco‐evolutionary consequences of interspecific phenological asynchrony – a theoretical perspective

The timing of biological events (phenology) is an important aspect of both a species’ life cycle and how it interacts with other species and its environment. Patterns of phenological change have been given much scientific attention, particularly recently in relation to climate change. For pairs of interacting species, if their rates of phenological change differ, then this may lead to asynchrony between them and disruption of their ecological interactions. However it is often difficult to interpret differential rates of phenological change and to predict their ecological and evolutionary consequences. We review theoretical results regarding this topic, with special emphasis on those arising from life history theory, evolutionary game theory and population dynamic models. Much ecological research on phenological change builds upon the concept of match/mismatch, so we start by putting forward a simple but general model that captures essential elements of this concept. We then systematically compare the predictions of this baseline model with expectations from theory in which additional ecological mechanisms and features of species life cycles are taken into account. We discuss the ways in which the fitness consequences of interspecific phenological asynchrony may be weak, strong, or idiosyncratic. We discuss theory showing that synchrony is not necessarily an expected evolutionary outcome, and how population densities are not necessarily maximized by adaptation, and the implications of these findings. By bringing together theoretical developments regarding the eco‐evolutionary consequences of phenological asynchrony, we provide an overview of available alternative hypotheses for interpreting empirical patterns as well as the starting point for the next generation of theory in this field.     

Rapid climate change and the rate of adaptation: insight from experimental quantitative genetics

CONTENTS: Summary 752 I. Introduction 752 II. Will migration be enough? 753 III. Can adaptation proceed fast enough? 754 IV. Fitness links demographic and evolutionary processes 755 V. Experimental studies: what do they tell us and how can we improve them? 756 VI. Predicting evolutionary change based on genetic variation and natural selection 757 VII. The chronosequence approach 758 VIII. Resurrection of ancestral propagules 759 IX. The mean and variance in fitness, a link between genetics and demography 760 X. Conclusions 762 Acknowledgements 762 References 762 SUMMARY: Evolution proceeds unceasingly in all biological populations. It is clear that climate-driven evolution has molded plants in deep time and within extant populations. However, it is less certain whether adaptive evolution can proceed sufficiently rapidly to maintain the fitness and demographic stability of populations subjected to exceptionally rapid contemporary climate change. Here, we consider this question, drawing on current evidence on the rate of plant range shifts and the potential for an adaptive evolutionary response. We emphasize advances in understanding based on theoretical studies that model interacting evolutionary processes, and we provide an overview of quantitative genetic approaches that can parameterize these models to provide more meaningful predictions of the dynamic interplay between genetics, demography and evolution. We outline further research that can clarify both the adaptive potential of plant populations as climate continues to change and the role played by ongoing adaptation in their persistence.     

Rapid evolution of escape ability in Trinidadian guppies (Poecilia reticulata)

Predators are widely assumed to create selection that shapes the evolution of prey escape abilities. However, this assumption is difficult to test directly due to the challenge of recording both predation and its evolutionary consequences in the wild. We examined these events by studying natural and experimental populations of Trinidadian guppies, Poecilia reticulata, which occur in distinct high-predation and low-predation environments within streams. Importantly, in the last two decades several populations of guppies have been experimentally introduced from one type of predatory environment into the other, allowing measurements of the consequences of change. We used this system to test two hypotheses: First, that changes in predatory environments create phenotypic selection favoring changes in escape ability of guppies, and second, that this selection can result in rapid evolution. For the first test we compared escape ability of wild caught guppies from high- versus low-predation environments by measuring survival rates during staged encounters with a major predator, the pike cichlid Crenicichla alta. We used guppies from three streams, comparing two within-stream pairs of natural populations and three within-stream pairs of an introduced population versus its natural source population. In every comparison, guppies from the high-predation population showed higher survival. These multiple, parallel divergences in guppy survival phenotype suggest that predatory environment does create selection of escape ability. We tested our second hypothesis by rearing guppies in common garden conditions in the laboratory, then repeating the earlier experiments using the F2 generation. As before, each comparison resulted in higher survival of guppies descended from the high-predation populations, demonstrating that population differences in escape ability have a genetic basis. These results also show that escape ability can evolve very rapidly in nature, that is, within 26-36 generations in the introduced populations. Interestingly, we found rapid evolutionary loss of escape ability in populations introduced into low-predation environments, suggesting that steep fitness trade-offs may influence the evolution of escape traits.     

Locally adapted traits maintained in the face of high gene flow

Gene flow between phenotypically divergent populations can disrupt local adaptation or, alternatively, may stimulate adaptive evolution by increasing genetic variation. We capitalised on historical Trinidadian guppy transplant experiments to test the phenotypic effects of increased gene flow caused by replicated introductions of adaptively divergent guppies, which were translocated from high‐ to low‐predation environments. We sampled two native populations prior to the onset of gene flow, six historic introduction sites, introduction sources and multiple downstream points in each basin. Extensive gene flow from introductions occurred in all streams, yet adaptive phenotypic divergence across a gradient in predation level was maintained. Descendants of guppies from a high‐predation source site showed high phenotypic similarity with native low‐predation guppies in as few as ~12 generations after gene flow, likely through a combination of adaptive evolution and phenotypic plasticity. Our results demonstrate that locally adapted phenotypes can be maintained despite extensive gene flow from divergent populations.     

Eco‐evolutionary partitioning metrics: assessing the importance of ecological and evolutionary contributions to population and community change

Interest in eco‐evolutionary dynamics is rapidly increasing thanks to ground‐breaking research indicating that evolution can occur rapidly and can alter the outcome of ecological processes. A key challenge in this sub‐discipline is establishing how important the contribution of evolutionary and ecological processes and their interactions are to observed shifts in population and community characteristics. Although a variety of metrics to separate and quantify the effects of evolutionary and ecological contributions to observed trait changes have been used, they often allocate fractions of observed changes to ecology and evolution in different ways. We used a mathematical and numerical comparison of two commonly used frameworks – the Price equation and reaction norms – to reveal that the Price equation cannot partition genetic from non‐genetic trait change within lineages, whereas the reaction norm approach cannot partition among‐ from within‐lineage trait change. We developed a new metric that combines the strengths of both Price‐based and reaction norm metrics, extended all metrics to analyse community change and also incorporated extinction and colonisation of species in these metrics. Depending on whether our new metric is applied to populations or communities, it can correctly separate intraspecific, interspecific, evolutionary, non‐evolutionary and interacting eco‐evolutionary contributions to trait change.     

Behavioural responses to human‐induced environmental change

The initial response of individuals to human‐induced environmental change is often behavioural. This can improve the performance of individuals under sudden, large‐scale perturbations and maintain viable populations. The response can also give additional time for genetic changes to arise and, hence, facilitate adaptation to new conditions. On the other hand, maladaptive responses, which reduce individual fitness, may occur when individuals encounter conditions that the population has not experienced during its evolutionary history, which can decrease population viability. A growing number of studies find human disturbances to induce behavioural responses, both directly and by altering factors that influence fitness. Common causes of behavioural responses are changes in the transmission of information, the concentration of endocrine disrupters, the availability of resources, the possibility of dispersal, and the abundance of interacting species. Frequent responses are alterations in habitat choice, movements, foraging, social behaviour and reproductive behaviour. Behavioural responses depend on the genetically determined reaction norm of the individuals, which evolves over generations. Populations first respond with individual behavioural plasticity, whereafter changes may arise through innovations and the social transmission of behavioural patterns within and across generations, and, finally, by evolution of the behavioural response over generations. Only a restricted number of species show behavioural adaptations that make them thrive in severely disturbed environments. Hence, rapid human‐induced disturbances often decrease the diversity of native species, while facilitating the spread of invasive species with highly plastic behaviours. Consequently, behavioural responses to human‐induced environmental change can have profound effects on the distribution, adaptation, speciation and extinction of populations and, hence, on biodiversity. A better understanding of the mechanisms of behavioural responses and their causes and consequences could improve our ability to predict the effects of human‐induced environmental change on individual species and on biodiversity.     

Loci under selection during multiple range expansions of an invasive plant are mostly population specific,but patterns are associated with climate

Identifying the genes underlying rapid evolutionary changes, describing their function and ascertaining the environmental pressures that determine fitness are the central elements needed for understanding of evolutionary processes and phenotypic changes that improve the fitness of populations. It has been hypothesized that rapid adaptive changes in new environments may contribute to the rapid spread and success of invasive plants and animals. As yet, studies of adaptation during invasion are scarce, as is knowledge of the genes underlying adaptation, especially in multiple replicated invasions. Here, we quantified how genotype frequencies change during invasions, resulting in rapid evolution of naturalized populations. We used six fully replicated common garden experiments in Brazil where Pinus taeda (loblolly pine) was introduced at the same time, in the same numbers, from the same seed sources, and has formed naturalized populations expanding outward from the plantations. We used a combination of nonparametric, population genetics and multivariate statistics to detect changes in genotype frequencies along each of the six naturalization gradients and their association with climate as well as shifts in allele frequencies compared to the source populations. Results show 25 genes with significant shifts in genotype frequencies. Six genes had shifts in more than one population. Climate explained 25% of the variation in the groups of genes under selection across all locations, but specific genes under strong selection during invasions did not show climate‐related convergence. In conclusion, we detected rapid evolutionary changes during invasive range expansions, but the particular gene‐level patterns of evolution may be population specific.     

Sex,competition and mimicry: an eco-evolutionary model reveals unexpected impacts of ecological interactions on the evolution of phenotypes in sympatry

Phenotypic evolution in sympatric species can be strongly impacted by species interactions, either mutualistic or antagonistic. Heterospecific reproductive behaviours between sympatric species have been shown to favour phenotypic divergence of traits used as sexual cues. Those traits may also be involved in local adaptation or in other types of species interactions and, as a result, undergo complex evolutions across sympatric species. Here we focus on mimicry and study how reproductive interference may impair phenotypic convergence between species with various levels of defence. We use a deterministic model assuming two sympatric species where individuals can display two different warning colour patterns. This eco-evolutionary model explores how ecological interactions shape phenotypic evolution within sympatric species. We investigate the effect of 1) the opposing density-dependent selections exerted on colour patterns by predation and reproductive behaviour and 2) the impact of relative species and phenotype abundances on the fitness costs faced by each individual depending on their species and phenotype. Our model shows that reproductive interference may limit the convergent effect of mimetic interactions and may promote phenotypic divergence between Müllerian mimics. The divergent and convergent evolution of traits also strongly depends on the relative species and phenotype abundances and levels of trophic competition, highlighting how the eco-evolutionary feedbacks between phenotypic evolution and species abundances may result in strikingly different evolutionary routes.     

CONSUMER–RESOURCE INTERACTIONS AND THE EVOLUTION OF MIGRATION

Theoretical studies have demonstrated that selection will favor increased migration when fitnesses vary both temporally and spatially, but it is far from clear how pervasive those theoretical conditions are in nature. Although consumer–resource interactions are omnipresent in nature and can generate spatial and temporal variation, it is unknown even in theory whether these dynamics favor the evolution of migration. We develop a mathematical model to address whether and how migration evolves when variability in fitness is determined at least in part by consumer–resource coevolutionary interactions. Our analyses show that such interactions can drive the evolution of migration in the resource, consumer, or both species and thus supplies a general explanation for the pervasiveness of migration. Over short time scales, we show the direction of change in migration rate is determined primarily by the state of local adaptation of the species involved: rates increase when a species is locally maladapted and decrease when locally adapted. Our results reveal that long‐term evolutionary trends in migration rates can differ dramatically depending on the strength or weakness of interspecific interactions and suggest an explanation for the evolutionary divergence of migration rates among interacting species.     

Increased juvenile predation is not associated with evolved differences in adult brain size in Trinidadian killifish (Rivulus hartii)

Vertebrates exhibit extensive variation in brain size. The long‐standing assumption is that this variation is driven by ecologically mediated selection. Recent work has shown that an increase in predator‐induced mortality is associated with evolved increases and decreases in brain size. Thus, the manner in which predators induce shifts in brain size remains unclear. Increased predation early in life is a key driver of many adult traits, including life‐history and behavioral traits. Such results foreshadow a connection between age‐specific mortality and selection on adult brain size. Trinidadian killifish, Rivulus hartii, are found in sites with and without guppies, Poecilia reticulata. The densities of Rivulus drop dramatically in sites with guppies because guppies prey upon juvenile Rivulus. Previous work has shown that guppy predation is associated with the evolution of adult life‐history traits in Rivulus. In this study, we compared second‐generation laboratory‐born Rivulus from sites with and without guppies for differences in brain size and associated trade‐offs between brain size and other components of fitness. Despite the large amount of existing research on the importance of early‐life events on the evolution of adult traits, and the role of predation on both behavior and brain size, we did not find an association between the presence of guppies and evolutionary shifts in Rivulus brain size. Such results argue that increased rates of juvenile mortality may not alter selection on adult brain size.     

Predicting range shifts under global change: the balance between local adaptation and dispersal

Bioclimate envelope models (BEMs) have often been criticized as being too simplistic due to e.g. not incorporating effects of biotic interactions or evolutionary adaptation. However, BEMs are widely applied and have proven to be often useful. Here we investigate, under which conditions evolution of dispersal, local adaptation or interspecific competition may be of minor importance for forecasting future range shifts. Therefore we use individual‐based simulations of metapopulations under climate change living in spatial temperature gradients. Scenarios incorporate single‐species systems or systems with competing species, respectively. Dispersal rate is evolving and adaptation to local conditions may also evolve in some scenarios. Results show that in single‐species scenarios excluding evolutionary adaptation, species either follow optimal habitat conditions or go extinct if habitat connectivity is too low. These simulations are in close accordance to predictions from BEMs. Including evolutionary adaptation qualitatively changes these results. In the absence of competing species the species either completely invades the world or goes extinct. With competitors, results strongly depend on habitat fragmentation. For highly connected habitats the range border may shift as predicted by BEMs, for intermediate connectivity it will lag behind, while species will go extinct if fragmentation is too high. Our results indicate that (simple) BEMs may work well if habitats are well connected and species will not encounter many difficulties in dispersing to new sites. Selection in this case may promote evolution of even higher dispersal activities. We thus show that the presence of biotic interactions may be ignored for predictions of range shifts when high dispersal can be expected.     

The implications of rapid eco‐evolutionary processes for biological control ‐ a review

Novel environmental conditions experienced by introduced species can drive rapid evolution of diverse traits. In turn, rapid evolution, both adaptive and non‐adaptive, can influence population size, growth rate, and other important ecological characteristics of populations. In addition, spatial evolutionary processes that arise from a combination of assortative mating between highly dispersive individuals at the expanding edge of populations and altered reproductive rates of those individuals can accelerate expansion speed. Growing experimental evidence shows that the effects of rapid evolution on ecological dynamics can be quite large, and thus it can affect establishment, persistence, and the distribution of populations. We review the experimental and theoretical literature on such eco‐evolutionary feedbacks and evaluate the implications of these processes for biological control. Experiments show that evolving populations can establish at higher rates and grow larger than non‐evolving populations. However, non‐adaptive processes, such as genetic drift and inbreeding depression can also lead to reduced fitness and declines in population size. Spatial evolutionary processes can increase spread rates and change the fitness of individuals at the expansion front. These examples demonstrate the power of eco‐evolutionary dynamics and indicate that evolution is likely more important in biocontrol programs than previously realized. We discuss how this knowledge can be used to enhance efficacy of biological control.     

Species interactions constrain adaptation and preserve ecological stability in an experimental microbial community

Species loss within a microbial community can increase resource availability and spur adaptive evolution. Environmental shifts that cause species loss or fluctuations in community composition are expected to become more common, so it is important to understand the evolutionary forces that shape the stability and function of the emergent community. Here we study experimental cultures of a simple, ecologically stable community of Saccharomyces cerevisiae and Lactobacillus plantarum, in order to understand how the presence or absence of a species impacts coexistence over evolutionary timescales. We found that evolution in coculture led to drastically altered evolutionary outcomes for L. plantarum, but not S. cerevisiae. Both monoculture- and co-culture-evolved L. plantarum evolved dozens of mutations over 925 generations of evolution, but only L. plantarum that had evolved in isolation from S. cerevisiae lost the capacity to coexist with S. cerevisiae. We find that the evolutionary loss of ecological stability corresponds with fitness differences between monoculture-evolved L. plantarum and S. cerevisiae and genetic changes that repeatedly evolve across the replicate populations of L. plantarum. This work shows how coevolution within a community can prevent destabilising evolution in individual species, thereby preserving ecological diversity and stability, despite rapid adaptation.Subject terms: Evolution, Microbial ecology     

Evolutionarily unstable fitness maxima and stable fitness minima of continuous traits

Summary We present models of adaptive change in continuous traits for the following situations: (1) adaptation of a single trait within a single population in which the fitness of a given individual depends on the population's mean trait value as well as its own trait value; (2) adaptation of two (or more) traits within a single population; (3) adaptation in two or more interacting species. We analyse a dynamic model of these adaptive scenarios in which the rate of change of the mean trait value is an increasing function of the fitness gradient (i.e. the rate of increase of individual fitness with the individual's trait value). Such models have been employed in evolutionary game theory and are often appropriate both for the evolution of quantitative genetic traits and for the behavioural adjustment of phenotypically plastic traits. The dynamics of the adaptation of several different ecologically important traits can result in characters that minimize individual fitness and can preclude evolution towards characters that maximize individual fitness. We discuss biological circumstances that are likely to produce such adaptive failures for situations involving foraging, predator avoidance, competition and coevolution. The results argue for greater attention to dynamical stability in models of the evolution of continuous traits.