The American Naturalist

AO_SCPLOWBSTRACTC_SCPLOWPredation often deviates from the law of mass action: many micro- and meso-scale experiments have shown that consumption saturates with resource abundance, and decreases due to interference between consumers. But does this observation hold at macro-ecological scales, spanning many species and orders of magnitude in biomass? If so, what are its consequences for large-scale ecological patterns and dynamics? We perform a meta-analysis of predator-prey pairs of mammals, birds and reptiles, and show that predation losses appear to increase, not as the product of predator and prey densities following the Lotka-Volterra (mass action) model, but rather as the square root of that product. This suggests a phenomenological power-law expression of the effective cross-ecosystem functional response. We discuss whether the same power-law may hold dynamically within an ecosystem, and assuming that it does, we explore its consequences in a simple food chain model. The empirical exponents fall close to the boundary between regimes of donor and consumer limitation. Exponents on this boundary are singular in multiple ways. First, they maximize predator abundance and some stability metrics. Second, they create proportionality relations between biomass and productivity, both within and between trophic levels. These intuitive relations do not hold in general in mass action models, yet they are widely observed empirically. These results provide evidence of mechanisms limiting predation across multiple ecological scales. Some of this evidence was previously associated with donor control, but we show that it supports a wider range of possibilities, including forms of consumer control. As limiting consumption counter-intuitively allows larger populations, it is worthwhile to reconsider whether the observed functional response arises from microscopic mechanisms, or could hint at selective pressure at the population level. This article has been peer-reviewed and recommended by Peer Community In Ecology (DOI: 10.24072/pci.ecology.100051)

In patch- or habitat-structured populations different processes can lead to diversity at different scales. While spatial heterogeneity generates spatially disruptive selection favoring variation between patches, local competition can lead to locally disruptive selection promoting variation within patches. So far, almost all theory has studied these two processes in isolation. Here, we use mathematical modelling to investigate how resource variation within and between habitats influences the evolution of variation in a consumer population where individuals compete in finite patches connected by dispersal. We find that locally and spatially disruptive selection typically act in concert, favoring polymorphism under a significantly wider range of conditions than when in isolation. But when patches are small and dispersal between them is low, kin competition inhibits the emergence of polymorphism, especially when driven by local competition. We further use our model to clarify what comparisons between trait and neutral genetic differentiation (Qst/Fst comparisons) can tell about the nature of selection. Overall, our results help understand the interaction between two major drivers of diversity: locally and spatially disruptive selection; and how this interaction is modulated by the unavoidable effects of kin selection under limited dispersal.

Global change is simplifying the structure of ecological networks; however, we are currently in a poor position to predict how these simplified communities will affect the evolutionary potential of remaining populations. Theory on adaptive landscapes provides a framework for predicting how selection constrains phenotypic evolution, but often treats the community context of evolving populations as a "black box". Here, we integrate ecological networks and adaptive landscapes to examine how changes in food-web complexity shape evolutionary constraints. We conducted a field experiment that manipulated the diversity of insect parasitoids (food-web complexity) that were able to impose selection on an insect herbivore. We then measured herbivore survival as a function of three key phenotypic traits. We found that more traits were under selection in simpler vs. more complex food webs. The adaptive landscape was more neutral in complex food webs because different parasitoid species impose different selection pressures, minimizing relative fitness differences among phenotypes. Our results suggest that phenotypic evolution becomes more constrained in simplified food webs. This indicates that the simplification of ecological communities may constrain the adaptive potential of remaining populations to future environmental change. "What escapes the eye, however, is a much more insidious kind of extinction: the extinction of ecological interactions." Janzen (1974)

Invasive species are valuable systems for evaluating evolutionary predictability, as populations in native and introduced ranges evolve separately, yet often encounter similar environmental challenges that select for parallel patterns of local adaptation. However, it remains unclear how pervasive and strong such parallelism is and how rapidly it evolves. To address these questions, we first extended cline theory to predict evolutionary patterns of parallelism between ranges. We then carried out a meta-analysis of clinal divergence in native and introduced populations of the same plant species and evaluated the tempo of evolutionary parallelism between ranges. Clines in introduced ranges were, on average, slightly shallower than native range clines for size, morpho-physiology, and phenology traits, but similar for reproductive and defense traits. Evolutionary parallelism of clinal divergence strongly increased with the time since introduction, with the greater parallelism in older introductions primarily caused by increased alignment in the direction of clinal divergence between ranges rather than changes in their relative magnitudes of divergence. These results are consistent with a two-phased process of cline evolution, in which introduced clines initially arise by drift during the range expansion, and subsequently evolve in response to local selection, ultimately leading to strong parallelism with the native range.

Demographic buffering and lability have been identified as adaptive strategies to optimise fitness in a fluctuating environment. These are not mutually exclusive, however we lack efficient methods to measure their relative importance for a given life history. Here, we decompose the stochastic growth rate (fitness) into components arising from nonlinear responses and variance-covariance of demographic parameters to an environmental driver, which allows studying joint effects of buffering and lability. We apply this decomposition for 154 animal matrix population models under different scenarios, to explore how these main fitness components vary across life histories. Faster-living species appear more responsive to environmental fluctuations, either positively or negatively. They have the highest potential for strong adaptive demographic lability, while demographic buffering is a main strategy in slow-living species. Our decomposition provides a comprehensive framework to study how organisms adapt to variability through buffering and lability, and to predict species responses to climate change.

With evolutionary rescue, a population that is declining due to an environmental change adapts to its environment, avoiding extinction. Previous theoretical work has explored the effects of negative density-dependence on rescue, showing that it may aid or hinder persistence. However, these models typically assume that it is only population intrinsic growth rates, r, or carrying capacity, K, that are negatively affected, and do not model density-dependence explicitly. Here, we analyze evolutionary rescue in a consumer-resource species following an abrupt environmental change, characterizing how rescue is dependent on the ecological effects of the environmental change on the consumer, which differently affect r and K through subsequent interactions with an explicit non-substitutable resource species. We derive approximate analytical predictions for the fixation probabilities of beneficial alleles, mutational supply, and times to extinction, which work well when selection is weak and individual turnover rates are low. We demonstrate that consumer rescue is dependent on the ecological effect of the environmental change, the resident life-history of the population, and the genetic architectures of evolving traits (monogenic versus polygenic). Our model suggests that measurements of intrinsic growth rates alone will be insufficient to predict rescue probabilities. This work extends our understanding of the interplay between ecology and evolution in influencing population persistence.

Coevolution can occur as a result of species interactions. However, it remains unclear how coevolutionary processes translate into the accumulation of species richness over macroevolutionary timescales. Assuming speciation occurs in a metacommunity as a result of genetic differentiation across communities due to dispersal limitation, we examine the effects of coevolution-induced stabilizing and destabilizing selection of a single quantitative trait on species diversification. We propose and test two hypotheses. (1) Stabilizing selection within communities enhances species diversification through strengthened dispersal limitation. (2) Destabilizing selection within communities impedes species diversification through weakened dispersal limitation. Here, we simulate clade co-diversification using an individual-based model, considering scenarios where phenotypic evolution is shaped by neutral dynamics, mutualistic coevolution, or antagonistic coevolution, where coevolution operates through trait matching or trait difference, and where the strength of coevolutionary selection is symmetrical or asymmetrical. Our assumption that interactions occur between an independent party (whose individuals can establish or persist in a community independently, e.g. hosts) and a dependent party (whose individuals cannot establish or persist in a community without the independent party, e.g. parasites or obligate mutualists) yields two contrasting results. Stabilizing selection within communities enhances species diversification in the dependent clade but not in the independent clade. Conversely, destabilizing selection within communities impedes species diversification in the independent clade but not in the dependent clade. These results are partially corroborated by empirical dispersal data, suggesting that these mechanisms might explain the diversification of some of the most species-rich clades in the Tree of Life.

Ecological specialization is widely thought to influence patterns of species richness by affecting rates at which species multiply and perish. Quantifying specialization is challenging, and using only one or a small number of ecological axes could bias estimates of overall specialization. Here, we calculate an index of specialization, based on seven measured traits, and estimate its effect on speciation and extinction rates in a large clade of birds. We find that speciation rate is independent of specialization, suggesting independence of local ecology and the geographic distributions of populations that promote allopatric species formation. Although some analyses suggest that more specialized species have higher extinction rates, leading to negative net diversification, this relationship is not consistently identified across our analyses. Our results suggest that specialization may drive diversification dynamics only on local scales or in specific clades, but is not generally responsible for macroevolutionary disparity in lineage diversification rates.

Gain curves were introduced to explain how hermaphrodites could displace a dioecious population, and to account for sexual allocation in hermaphrodites. Terms for gamete production employed for the first purpose were transformed for the second into male and female gain curves that ostensibly defined fitness outcomes. These gain curves pose a conceptual challenge if they are specified separately because fitness at the population level cannot occur through one sex function independently of success through the other. If gain curves truly represent fitness outcomes, anomalies can arise, such as inequality of total male and female fitness in a population. Gain curves were originally used in a mathematical framework that treated the ostensible gain functions as inputs of male and female actors to a mating arena rather than as mating outcomes from that arena. I present a model of sex allocation that incorporates power functions to describe both gamete production and fitness gain in a manner that explicitly separates these two roles. In this formulation, the gamete production functions have the identical effect on optimal sex allocation originally attributed to gain curves while the true fitness gain curves lose nearly all effect on the optimum. Thus, despite the label, gain curves were implicitly describing inputs rather than outcomes. Because gain curves have been a staple of evolutionary ecology for decades, the implication is that much of our understanding of sexual allocation in hermaphrodites needs to be revisited. I outline some directions such an effort might take.

Every species experiences limits to its geographic distribution. Some evolutionary models predict that populations at range edges are less well-adapted to their local environments due to drift, expansion load, or swamping gene flow from the range interior. Alternatively, populations near range edges might be uniquely adapted to marginal environments. In this study, we use a database of transplant studies that quantify performance at broad geographic scales to test how local adaptation, site quality, and population quality change from spatial and climatic range centers towards edges. We find that populations from poleward edges perform relatively poorly, both on average across all sites (15% lower population quality) and when compared to other populations at home (31% relative fitness disadvantage), consistent with these populations harboring high genetic load. Populations from equatorial edges also perform poorly on average (18% lower population quality) but, in contrast, outperform foreign populations (16% relative fitness advantage), suggesting that populations from equatorial edges have strongly adapted to unique environments. Finally, we find that populations from sites that are thermally extreme relative to the species niche demonstrate strong local adaptation, regardless of their geographic position. Our findings indicate that both nonadaptive processes and adaptive evolution contribute to variation in adaptation across species ranges.

When two species use the same resource, this typically leads to competition, such as when different plants aim to attract the same mutualist pollinators. However, more flowers may also attract more pollinators to an area, such that one or both competitors actually benefit from the others presence. For example, it has been argued that strips of wildflowers planted next to crops may attract pollinators who spill over into the crop. Here we mathematically examine facilitation and competition in consumer attraction. Contrary to previous claims, no accelerating benefits of density per se are necessary for facilitation. Instead, under very general assumptions, facilitation can be generated by an imbalance between local competition and joint long-distance attraction of consumers; for example, a low presence of highly attractive wildflowers should lead to benefits to a crop. In this mechanism, how pollinator attraction to a patch increases with density of plants is a key factor. Our results generalize to many contexts where local competition may trade off with joint long-distance attraction of consumers, and we show that the exact relationship between competitor density and attraction of consumers can qualitatively shape outcomes, including facilitation or competition.

Partial clonality is widespread in natural populations; however, the distributions of clone ages under different rates of clonality and their effects on genetic diversity remain unexplored. To fill this gap, we simulated partially clonal populations over 10,000 generations across a range of clonality and mutation rates. Using a forward-in-time individual-based model, we evaluated genetic and genotypic indices alongside measures of clone age distribution to examine how different clonality rates influence clone age distributions, and how these distributions impact population structure and genetic diversity over time. Our results reveal two distinct trajectories: (i) at low to moderate rates of clonality evolution is driven by sharp and predictable patterns of rapid clone turnovers, resulting in predominantly young clones which relative abundances well align with genotypic indices ; (ii) at extreme rates of clonality, demographic stochasticity generates very variable clone age distributions which mirror the high variance of mean and variance of FIS that summarize gene reshuffling between individuals. Interestingly, a portion of the variability of these indices can be explained by differences in clone age distribution that occur by chance. Our results, therefore, complement previous theoretical studies on the population genetics of partial clonality by providing biological insights into how clonal turnover dynamics shape the temporal evolution and variability of highly clonal populations. These results have practical implications for inferring evolutionary trajectories and managing partially clonal species.

A fundamental feature of pollination systems is the indirect facilitation and competition that arises when plants species share pollinators. When plants share pollinators, the pollination service can be influenced. This depends not only on how many partners plant species share, but also by multiple intertwined factors like the plant species abundance, visitation, or traits. These factors inherently operate at the community level. However, most of our understanding of how these factors may affect the pollination service is based on systems of up to a handful of species. By examining comprehensive empirical data in eleven natural communities, we show here that the pollination service is--surprisingly--only partially influenced by the number of shared pollinators. Instead, the factors that most influence the pollination service (abundance and visit effectiveness) also introduce a trade-off between the absolute amount of conspecific pollen received and the amount relative to heterospecific pollen. Importantly, the ways plants appear to balance these trade-offs depend strongly on the community context, as most species showed flexibility in the strategy they used to cope with competition for pollination.

Dispersal is one of the most important drivers of community assembly. The conventional belief that dispersal leads to biotic homogenization (lower beta diversity) has been recently challenged by an experiment conducted in nectar microbes (Vannette & Fukami, 2017), showing that dispersal could lead to community divergence. In this paper, I re-examined the relationship between beta diversity and local dispersal in a range of theoretical models: from the classic island biogeography model and meta-population model to a meta-community model that incorporates biotic interactions. I find that the emergence of hump-shaped beta diversity-dispersal relationship is closely related to local dispersal (rather than global dispersal), non-neutrality and biotic interactions. The results reveal rich metacommunity dynamics in relation to dispersal types and biotic interactions which might be overlooked in previous theoretical and empirical studies. The findings call for more realistic experimental manipulations on dispersals in future community assembly studies.

There is increasing evidence that life-history traits can evolve rapidly during range expansion and that this evolution can impact the ecological dynamics of population spread. While dispersal evolution during range expansion has received substantial attention, dormancy (dispersal in time) has not. Here, we use an individual-based model to investigate the evolution of seed dormancy during range expansion. When a population is at spatial equilibrium our model produces results that are consistent with previous theoretical studies: seed dormancy evolves due to kin competition and the degree of dormancy increases as temporal environmental variation increases. During range expansions we consistently observe evolution towards reduced rates of dormancy at the front. Behind the front there is selection for higher rates of dormancy. Notably, the decreased dormancy towards the expanding margin reduces the regional resilience of recently expanded populations to a series of harsh years. We discuss how dormancy evolution during range expansion, and its consequences for spatial population dynamics, may impact other evolutionary responses to environmental change. We end with suggestions for future theoretical and empirical work.

Biodiversity results from differentiation mechanisms developing within biological populations. Such mechanisms are influenced by the properties of the landscape over which individuals interact, disperse and evolve. Notably, landscape connectivity and habitat heterogeneity constrain the movement and survival of individuals, thereby promoting differentiation through drift and local adaptation. Nevertheless, the complexity of landscape features can blur our understanding of how they drive differentiation. Here, we formulate a stochastic, eco-evolutionary model where individuals are structured over a graph that captures complex connectivity patterns and accounts for habitat heterogeneity. Individuals possess neutral and adaptive traits, whose divergence results in differentiation at the population level. The modelling framework enables an analytical underpinning of emerging macroscopic properties, which we complement with numerical simulations to investigate how the graph topology and the spatial habitat distribution affect differentiation. We show that in the absence of selection, graphs with high characteristic length and high heterogeneity in degree promote neutral differentiation. Habitat assortativity, a metric that captures habitat spatial autocorrelation in graphs, additionally drives differentiation patterns under habitat-dependent selection. While assortativity systematically amplifies adaptive differentiation, it can foster or depress neutral differentiation depending on the migration regime. By formalising the eco-evolutionary and spatial dynamics of biological populations in complex landscapes, our study establishes the link between landscape features and the emergence of diversification, contributing to a fundamental understanding of the origin of biodiversity gradients. Significance statementIt is not clear how landscape connectivity and habitat heterogeneity influence differentiation in biological populations. To obtain a mechanistic understanding of underlying processes, we construct an individualbased model that accounts for eco-evolutionary and spatial dynamics over graphs. Individuals possess both neutral and adaptive traits, whose co-evolution results in differentiation at the population level. In agreement with empirical studies, we show that characteristic length, heterogeneity in degree and habitat assortativity drive differentiation. By using analytical tools that permit a macroscopic description of the dynamics, we further link differentiation patterns to the mechanisms that generate them. Our study provides support for a mechanistic understanding of how landscape features affect diversification.

Mutualistic species vary in their level of partner specificity, which has important evolutionary, ecological, and management implications. Yet, the evolutionary mechanisms which underpin partner specificity are not fully understood. Most work on specialization focuses on the trade-off between generalism and specialism, where specialists receive more benefits from preferred partners at the expense of benefits from non-preferred partners, while generalists receive similar benefits from all partners. Because all mutualisms involve some degree of both cooperation and conflict between partners, we highlight that specialization to a mutualistic partner can be cooperative, increasing benefit to a focal species and a partner, or antagonistic, increasing resource extraction by a focal species from a partner. We devise an evolutionary game theoretic model to assess the evolutionary dynamics of cooperative specialization, antagonistic specialization, and generalism. Our model shows that cooperative specialization leads to bistability: stable equilibria with a specialist host and its preferred partner excluding all others. We also show that under cooperative specialization with spatial effects, generalists can thrive at the boundaries between differing specialist patches. Under antagonistic specialization, generalism is evolutionarily stable. We provide predictions for how a cooperation-antagonism continuum may determine the patterns of partner specificity that develop within mutualistic relationships.

Bet-hedging strategies, such as dispersal and dormancy, are predicted to evolve in varying and uncertain environments and are critical to ecological models of biodiversity maintenance. Theories of the specific ecological scenarios that favor the evolution of dispersal, dormancy, or their covariance are rarely tested, particularly for naturally-evolved populations that experience complex patterns of spatiotemporal environmental variation. We grew 23 populations of Vulpia microstachys, an annual grass native to California, in a greenhouse, and on the offspring generation measured seed dispersal ability and dormancy rates. We hypothesized that seed dormancy rates and dispersal abilities would be highest in populations from more productive, temporally variable sites, causing them to covary positively. Contrary to our hypothesis, our data suggest that both dispersal and dormancy evolve to combat different axes and scales of spatial heterogeneity, and are underlain by different seed traits, allowing them to evolve independently. Dormancy appears to have evolved as a strategy for overcoming microgeographic heterogeneity rather than temporal climate fluctuations, an outcome that to our knowledge has not been considered by theory. In sum, we provide much needed empirical data on the evolution of bet hedging, as well as a new perspective on the ecological function dormancy provides in heterogeneous landscapes.

Many generalist species consist of disparate specialized individuals, a phenomenon known as individual specialization. This within-population niche variation can stabilize population dynamics, reduce extinction risk, and alter community composition. But, we still only vaguely understand the ecological contexts that promote niche variation and its stabilizing effects. Adaptive dynamics models predict that intraspecific variation should be greater in environments with two or more equally-profitable resources, but reduced in environments dominated by one resource. Here, we confirm this prediction using a comparison of threespine stickleback in 33 lakes in on Vancouver Island, Canada. Stickleback consume a combination of benthic and limnetic invertebrates, focusing on the former in small lakes, the latter in large lakes. Intermediate-sized lakes support generalist populations, which arise via greater among-individual diet variation, not by greater individual diet breadth. These intermediate lakes exhibit correspondingly greater morphological diversity, while genomic diversity increases linearly with lake size. These results support the theoretical expectation that habitats with an intermediate ratio of resources are \"just right\" for promoting ecologically relevant intraspecific diversification.

The ecology of fear demonstrates how prey responses to avoid predation cause non-lethal effects at all ecological scales. Parasites also elicit defensive responses in hosts with associated non-lethal effects, which raises the longstanding, yet unresolved question of how non-lethal effects of parasites compare with those of predators. We developed a framework for systematically answering this question for all types of predator and parasite systems. Our framework predicts that trait responses and their non-lethal effects should be strongest from predators and parasites that do not kill individuals to feed on them, but which nevertheless damage fitness. Analysing trait response data on amphibians, which have been well-studied for this area of research, showed that individuals generally responded more directly to short-term predation risks than to parasitism. Apart from studies using amphibians, there have been few direct comparisons of responses to predation and parasitism, and none have incorporated responses to micropredators, parasitoids, or parasitic castrators, or examined their long-term consequences. Addressing these and other data gaps highlighted by our general framework can advance the field toward understanding how non-lethal effects shape real food webs, which contain multiple predator and parasite species.