AbstractExtreme cold events, which have become more frequent, can revert the direction of long-term responses to climate change. In 2021, record snowstorms swept the United States, causing wildlife die-offs that may have been associated with rapid natural selection. Our objective was to determine whether the snowstorms caused natural selection in Eastern Bluebirds (). To test which mechanism most influenced their survival, we measured the morphology and coloration of fatalities and survivors at three sites. Survival was associated with a longer tarsus and with a wider, longer, and deeper beak, in support of the starvation and thermal endurance hypotheses. Additionally, bluebirds with more-ornamented plumage were less likely to have survived, perhaps because of an early energy investment in mate and site acquisition. As bluebirds encounter increasingly warm summer conditions, the longer extremities favored during the snowstorms may continue to be favored through their thermoregulatory benefits. However, the dull plumage coloration favored by natural selection during the snowstorms may be opposed by sexual selection benefits of more-ornamented plumage. Overall, responses to extreme events are difficult to predict from responses to long-term climate change, and responses to one event, such as the 2021 snowstorms, may not predict responses to a future extreme event.

AbstractHow do species' distributions respond to their environments? This question was at the heart of the Clements-Gleason controversy, ecology's most famous debate. Do species respond to the environment in concerted ways, leading to distinct and cohesive assemblages (the Clementsian paradigm), or do species respond to the environment independently (the Gleasonian paradigm)? Using plant occurrences along the elevation gradient of Pikes Peak (Colorado) as a lens through which to gain insight into Clements's perspectives on the debate, we formally test for community patterns along this gradient using a modern framework unavailable at the time of Clements and Gleason. The Pikes Peak region was Clements's study area for more than 40 years, where he established a research lab and distributed sites along the elevational gradient. His investigations of plant distributions on this mountain likely influenced his views on communities. We found mixed support for the paradigms, with neither the Gleasonian paradigm nor the Clementsian paradigm fully supported. While distributions along the gradient showed evidence of clustering of species range edges, considered to be consistent with the Clementsian paradigm, the pattern was weak, and neither range edges nor species turnover peaked at ecotone elevations, as expected under the Clementsian paradigm. Our results illuminate the Clements-Gleason debate by allowing us to probe issues that complicate conclusively testing the paradigms, such as deciding on how we quantify environmental gradients and determining the appropriate scales for community patterns and processes that might generate them. Revisiting the debate also revealed that Clements's and Gleason's views had more in common than we realize. The debate may be less neatly resolved than we assume from mythos, and it continues to have relevance to basic and applied ecology today, as its legacy has shaped our (still tenuous) notion of ecological communities and the trajectory of our field.

AbstractHow bees shift the timing of their seasonal activity (phenology) to track favorable conditions influences the degree to which bee foraging and flowering plant reproduction overlap. While bee phenology is known to shift due to interannual climatic variation and experimental temperature manipulation, the underlying causes of these shifts are poorly understood. Most studies of bee phenology have been phenomenological and have only examined shifts of point estimates, such as first appearance or peak timing. Such cross-sectional measures are convenient for analysis, but foraging activity is distributed across time, and pollination interactions are better described by overlap in phenological abundance curves. Here, we make simultaneous inferences about interannual shifts in bee phenology, emergence and senescence rates, population size, and the effect of floral abundance on observed bee abundance. We do this with a model of transition rates between life stages implemented in a hierarchical Bayesian framework and parameterized with fine-scale abundance time series of the sweat bee at the Rocky Mountain Biological Laboratory in Colorado. We find that 's emergence cueing was highly sensitive to the timing of snowmelt but that emergence rate, senescence rate, and population size did not differ greatly across years. The present approach can be used to glean information about vital rates from other datasets on bee and flower phenology, improving our understanding of pollination interactions.

AbstractMany parasite species use multiple host species to complete development; however, empirical tests of models that seek to understand factors impacting evolutionary changes or maintenance of host number in parasite life cycles are scarce. Specifically, one model incorporating parasite mating systems that posits that multihost life cycles are an adaptation to prevent inbreeding in hermaphroditic parasites and thus preclude inbreeding depression remains untested. The model assumes that loss of a host results in parasite inbreeding and predicts that host loss can evolve only if there is no parasite inbreeding depression. We provide the first empirical tests of this model using a novel approach we developed for assessing inbreeding depression from field-collected parasite samples. The method compares genetically based selfing rate estimates to a demographic-based selfing rate, which was derived from the closed mating system experienced by endoparasites. Results from the hermaphroditic trematode , which has a derived two-host life cycle, supported both the assumption and the prediction of the mating system model, as this highly inbred species had no indication of inbreeding depression. Additionally, comparisons of genetic and demographic selfing rates revealed a mixed mating system that could be explained completely by the parasite's demography (i.e., its infection intensities).

AbstractFire events change background color, impairing camouflage strategies. However, selection for polymorphic populations may increase camouflage and survival by reducing predation risks. We conducted experiments addressing background selection and predation pressures on the effectiveness of arthropod camouflage against predation in burned and unburned trunks. We tested color and luminance contrasts, as well as trunk preferences, in a color polymorphic grasshopper and praying mantis species with melanic and brown morphs, and a spider species with a single dark color. To expand the scope of our study, we used two distinct visual models of avian predators: ultraviolet sensitive and violet sensitive. We also performed predation experiments using theoretical prey exhibiting black and brown color and human "predators" to understand the effectiveness of color polymorphism against different trunk conditions. Melanic morphs had lower achromatic contrast in burned backgrounds for both visual systems, suggesting that melanism promotes advantages against predation over long distances. However, only spiders actively selected the low-contrasting burned trunks, indicating habitat specialization. The predation experiments showed that black models benefited from camouflage on burned trunks. Conversely, brown models elicited more time and reduced distance in predator searching compared with the black targets on unburned trunks. We suggest that postfire effects can enhance color contrasts and increase predation over color-mismatching individuals, which translates into selection pressures for color polymorphism and matching background choices.

AbstractPollen grains from different plants potentially compete for ovule access because flowers produce many more pollen grains than ovules. Pollen competition could occur on pollinators, where there is finite space for pollen placement. Here, we explore the explosive pollen deposition in (Lamiaceae, a perennial flowering plant native to South America that is frequently visited by hummingbirds) and determine whether it can improve male performance by reducing pollen loads deposited by previously visited flowers. Through the simulation of floral visits utilizing a hummingbird skull, we showed that explosive pollen deposition by untriggered flowers dislodges almost twice as many pollen grains as already-triggered flowers. In addition, pollen removal increases with the amount of deposited pollen by the floral explosion, suggesting that the precision or the explosive force of pollen deposition plays a pivotal role in this pollen removal process. These results suggest that explosive pollen placement, a mechanism that has evolved in many unrelated angiosperm clades, may confer a prepollination male competition advantage to plants.

AbstractRandom search theories predict that animals employ movement patterns that optimize encounter rates with target resources. However, animals are not always able to achieve the best search strategy. Energy depletion, for example, limits searchers' movement activities, forcing them to adjust their behaviors before and after encounters. Here, we investigate the cost of mate search in a termite, , and reveal that the costs associated with mate finding reduce the selectivity of mating partners. After a dispersal flight, termites search for a mating partner with limited reserved energy. We found that their movement activity and diffusiveness progressively declined over extended mate search. Our data-based simulations qualitatively confirmed that the reduced movement diffusiveness decreased the searching efficiency. Also, prolonged search periods reduced survival rate and the number of offspring. Thus, mate search has two different negative effects on termites. Finally, we found that termites with an extended mate search reduced the selectivity of mating partners, where males immediately paired with any encountering females. Thus, termites dramatically changed their mate search behavior depending on their internal states. Our finding highlights that accounting for the searchers' internal states is essential to fill the gap between random search theories and empirical behavioral observations.

AbstractDensity dependence is often assumed in population dynamics, but its importance in small, isolated populations has been questioned. We evaluated the relative influence of density dependence, environmental conditions, and sporadic events (disease outbreaks and specialist predators) on annual population growth rate, annual female reproduction, and annual survival of juveniles and adult females in three populations of mountain ungulates. We analyzed long-term (30-47 years) individual-based data on two bighorn sheep populations and one mountain goat population in Alberta, Canada. The effect of cougar predation episodes and pneumonia epizootics on annual population growth rate was twice as strong as that of population density. While pneumonia reduced adult female and juvenile survival and predation episodes decreased all demographic rates, high density lowered only juvenile survival. Long-term studies are pivotal for understanding the dynamics of large herbivore populations, but they are rarely duplicated. Our analysis of three mountain ungulate populations with similar life history and ecological characteristics provides evidence that infrequent sporadic events can have a greater relative influence on annual population growth than density-dependent factors in isolated populations. This result contrasts with studies of larger, well-connected populations, highlighting the importance of considering sporadic events in the management and conservation of isolated populations.

AbstractNonliving resources frequently flow across ecosystem boundaries, which can yield networks of spatially coupled ecosystems. Yet the significance of resource flows for ecosystem function has predominantly been understood by studying two or a few coupled ecosystems, overlooking the broader resource flow network and its spatial structure. Here, we investigate how the spatial structure of larger resource flow networks influences ecosystem function at metaecosystem scales by analyzing metaecosystem models with homogeneously versus heterogeneously distributed resource flow networks but otherwise identical characteristics. We show that metaecosystem function can differ strongly between metaecosystems with contrasting resource flow networks. Differences in function generally arise through the scaling up of nonlinear local processes interacting with spatial variation in local dynamics, the latter of which is influenced by network structure. However, we find that neither network structure guarantees the greatest metaecosystem function. Rather, biotic (organism traits) and abiotic (resource flow rates) properties interact with network structure to determine which yields greater metaecosystem function. Our findings suggest that the spatial structure of resource flow networks coupling ecosystems can be a driver of ecosystem function at landscape scales. Furthermore, our study demonstrates how modifications to the structural, biotic, or abiotic properties of metaecosystem networks can have nontrivial large-scale effects on ecosystem function.

AbstractEnvironmental conditions (i.e., climatic variation) can strongly influence the cost and benefits of reproductive traits. Yet there is still no consensus on whether changing environmental conditions strengthen or relax sexual selection. Evidence from the literature suggests that highly variable environments can limit mate choice and investment in sexual traits, hence relaxing sexual selection pressures. Here, we tested this hypothesis using the nuptial gift-giving spider , in which males can either wrap nutritive (fresh prey) or worthless (prey leftovers) items in silk. We examined changes in males' sexual trait and female choice among six populations living under different climatic conditions. We found that large variation in precipitation limits female choice, potentially favoring the spread of deceptive worthless gifts. In populations under highly variable conditions and with the highest frequencies of worthless gifts (70%), males offering such gifts acquire longer mating durations than those offering nutritive gifts. In contrast, in populations with less variable conditions and the lowest frequencies of worthless gift (36%), females shortened mating duration to males offering worthless gifts. Our findings are consistent with the prediction that highly variable environmental conditions relax sexual selection.

AbstractHormones mediate sexual dimorphism by regulating sex-specific patterns of gene expression, but it is unclear how much of this regulation involves sex-specific hormone levels versus sex-specific transcriptomic responses to the same hormonal signal. Moreover, transcriptomic responses to hormones can evolve, but the extent to which hormonal pleiotropy in gene regulation is conserved across closely related species is not well understood. We addressed these issues by elevating testosterone levels in juvenile females and males of three lizard species before sexual divergence in circulating testosterone and then characterizing transcriptomic responses in the liver. In each species, more genes were responsive to testosterone in males than in females, suggesting that early developmental processes prime sex-specific transcriptomic responses to testosterone later in life. However, overall transcriptomic responses to testosterone were concordant between sexes, with no genes exhibiting sex-by-treatment interactions. By contrast, hundreds of genes exhibited species-by-treatment interactions, particularly when comparing distantly related species with different patterns of sexual dimorphism, suggesting evolutionary lability in gene regulation by testosterone. Collectively, our results indicate that early organizational effects may lead to sex-specific differences in the magnitude, but not the direction, of transcriptomic responses to testosterone and that the hormone-genome interface accrues regulatory changes over evolutionary time.

AbstractFor species that partition resources, the classic expectation is that increasing resource diversity allows for increased species diversity. On the other hand, for neutral species, such as those competing equally for a single resource, diversity reflects a balance between the rate of introduction of novelty (e.g., by immigration or speciation) and the rate of extinction. Recent models of microbial metabolism have identified scenarios where metabolic trade-offs among species partitioning multiple resources can produce emergent neutral-like dynamics. In this hybrid scenario, one might expect that both resource diversity and immigration will act to boost species diversity. We show, however, that the reverse may be true: when metabolic trade-offs hold and population sizes are sufficiently large, increasing resource diversity can act to reduce species diversity, sometimes drastically. This reversal is explained by a generic transition between neutral- and niche-like dynamics, driven by the diversity of resources. The inverted resource-diversity relationship that results may be a signature of consumer-resource systems with strong metabolic trade-offs.

AbstractSocial environments impose a number of constraints on individuals' behavior. These constraints have been hypothesized to generate behavioral variation among individuals, social responsiveness, and within-individual behavioral consistency (also termed "predictability"). In particular, the social niche specialization hypothesis posits that higher levels of competition associated with higher population density should increase among-individual behavioral variation and individual predictability as a way to reduce conflicts. Being predictable should hence have fitness benefits in group-living animals. However, to date empirical studies of the fitness consequences of behavioral predictability remain scarce. In this study, we investigated the associations between social behavior, its predictability, and fitness in the eastern water dragon (), a wild gregarious lizard. Since this species is sexually dimorphic, we examined these patterns both between sexes and among individuals. Although females were more sociable than males, there was no evidence for sex differences in among-individual variation or predictability. However, females exhibited positive associations between social behavior, its predictability, and survival, while males exhibited only a positive association between mean social behavior and fitness. These findings hence partly support predictions from the social niche specialization hypothesis and suggest that the function of social predictability may be sex dependent.

AbstractThe widespread occurrence of genetically programmed cell death (PCD) in unicellular species poses an evolutionary puzzle. While kin selection theory predicts that the fitness benefits of cell suicide must be preferentially directed toward genetic relatives, it does not predict the nature of these benefits. Furthermore, cell suicide must be conditionally expressed, leaving open the question of what conditions optimally regulate expression. Here we formalize several verbal hypotheses for the ecological function of unicellular PCD. We show that self-sacrifice by healthy cells cannot evolve. Instead, PCD evolution requires that damaged cells sense impending death and then (1) expedite this death to spare resources for groupmates, (2) prepare cellular contents so that necrotic toxins are not released upon death, or initiate autolysis in order to (3) release beneficial compounds or (4) release anticompetitior toxins. The prerequisite ability to predict death is a severe cell biological constraint as well as an ecological constraint that restricts PCD evolution to species with specific sources of mortality. We show that the specific type of PCD that will evolve, though, differs on the basis of a species' ecology, life history, and genetic structure.

AbstractIndirect genetic effects (IGEs) exist when there is heritable variation in one organism's ability to alter a second organism's traits. For example, parasites have antigens that can induce a host immune response, as well as disparate strategies to evade or suppress host immunity; among-parasite genetic variation in these antigens generates among-host variation in immune traits. Here, we experimentally show that the cestode parasite exerts an IGE on an immune trait (peritoneal fibrosis) in its threespine stickleback host: stickleback developed strong fibrosis after exposure to some parasite genotypes but not others. A complication arises during coinfection, when two or more parasite genotypes may impose conflicting IGEs on the same host trait. What parasite-controlled trait will the host express? Will the host trait reflect the more immune-stimulatory parasite genotype or the more immune-evasive genotype? These alternatives can be quantified by estimating the dominance coefficient, as if a coinfected host were a heterozygote. We experimentally estimated the dominance of IGEs by coinjecting antigens from different parasite genotypes. Contrary to our a priori hypotheses, coinjected antigens induced an overdominant effect, stronger than either parasite's antigens alone. We present a mathematical model showing that the value of this IGE dominance is biologically important, affecting the evolutionary dynamics of parasites in a density- and frequency-dependent manner. The model indicates that overdominance would be detrimental to immigrants when resident prevalence is high. This combination of experimental data and modeling provides an example of a parasite IGE on host traits and the evolutionary significance of IGE dominance.

AbstractThe matrix is the matrix of additive genetic variances and covariances for a vector of phenotypes. Here we apply the classical theory for the balance among selection, genetic drift, and mutations to find the contributions to from each locus for populations at stasis. The fitness is approximated by a linear function of phenotypes, with coefficients affected by environmental fluctuations. We show that the matrix can be decomposed into four additive components generated by selection, drift, mutations, and environmental fluctuations. Selection is on average counteracted by the other three processes included in Fisher's concept of deterioration of the environment, generating considerable changes in mean phenotypes. The theory illustrates that neither Fisher's fundamental theorem nor Lande's classical gradient formula is sufficient for assessing adaptive changes through time unless the deteriorations are corrected for. This applies for populations at stasis, but also for populations that are subject to long-term evolutionary changes. The theory also indicates several possible comparative studies for investigations of deteriorating effects. Our analyses also suggest that the factor loadings to the eigenvector of the matrix with the lowest eigenvalue will rather accurately indicate the relative contributions from different phenotype components to fitness. This is information notoriously difficult to obtain in natural populations.

AbstractUnderstanding patterns of diversification necessarily requires accounting for both the generation and the persistence of species. Formal models of speciation genetics, however, focus on the generation of new species without explicitly considering the maintenance of biodiversity (e.g., coexistence, the focus of ecological studies of diversity). Consequently, it remains unclear whether and how new species will coexist following a speciation event, a gap limiting our ability to understand the rate-limiting controls of diversification over macroevolutionary timescales. To connect coexistence and speciation theory and assess the relative importance of ecological versus genetic constraints in diversification events, we develop a deterministic, three-locus, population-genetic model that includes a skewed distribution of available resources (to generate variation in fitness differences), frequency-dependent competition, and assortative mating. Both ecology and genetics play vital and interacting roles in shaping initial speciation events and long-term eco-evolutionary outcomes. Ecological constraints are especially important when fitness differences are large and competition remains strong among dissimilar phenotypes. Ephemeral species can occur in our model and are typically lost because of competitive exclusion, a result demonstrating that species persistence may serve as the rate-limiting control of long-term diversification rates. More broadly, our model adds evidence that the unification of ecological and evolutionary (including genetic) perspectives on biodiversity is needed to predict large-scale patterns.

AbstractHost organisms can harbor microbial symbionts that defend them from pathogen infection in addition to the resistance encoded by the host genome. Here, we investigated how variation in defenses, generated from host genetic background and symbiont presence, affects the emergence of pathogen genetic diversity across evolutionary time. We passaged the opportunistic pathogen through populations of the nematode varying in genetic-based defenses and prevalence of a protective symbiont. After 14 passages, we assessed the amount of genetic variation accumulated in evolved pathogen lineages. We found that diversity begets diversity. An overall greater level of pathogen whole-genome and per-gene genetic diversity was measured in pathogens evolved in mixed host populations compared with those evolved in host populations composed of one type of defense. Our findings directly demonstrate that symbiont-generated heterogeneity in host defense can be a significant contributor to pathogen genetic variation.

AbstractA central challenge in community ecology is understanding and predicting the effects of abiotic factors on community assembly. In particular, microbial communities play a central role in the ecosystem, but we do not understand how changing factors like temperature are going to affect community composition or function. In this article, we studied the self-assembly of multiple communities in synthetic environments to understand changes in microbial community composition based on metabolic responses of different functional groups along a temperature gradient. In many microbial communities, different microbial functional groups coexist through the partitioning of carbon sources in an emergent trophic structure (cross-feeding). In this system, respirofermentative bacteria display a preference for the sugars supplied as the only carbon source but secrete secondary carbon sources (organic acids) that are more efficiently consumed by obligate respirators. As a consequence of this trophic structure, the metabolic plasticity of the respirofermenters has downstream consequences for the relative abundance of respirators across temperatures. We found that the effects of different temperatures on microbial composition can largely be described by an increase in fermentation by-products with increasing temperatures from the respirofermentative bacteria. This research highlights the importance of metabolic plasticity and metabolic trade-offs in predicting species interactions and community dynamics across abiotic gradients.