Natural enemies are ubiquitous in nature. In many communities, natural enemies have a major effect on the diversity of their prey. Their effects are very diverse: they can promote or undermine the ability of their prey to coexist through a variety of mechanisms. As such, an important step in understanding how diversity is maintained will be to understand how different forms of predator behavior affect prey coexistence. In this dissertation, I study how two major types of predators affect plant coexistence in two different communities. First, I study natural enemies in tropical forests, using both theory and empirical work. In tropical forests, most natural enemies are thought have a narrow host range, and be distance-responsive (i.e., mainly harm seeds and seedlings that are near adults of their main host). Previous theoretical work has shown that specialized natural enemies can maintain diversity of their prey, whether or not they are distance-responsive. However, it is unknown whether specialist natural enemies are more or less able to promote prey coexistence if they are distance-responsive. Using theoretical models, I show that distance-responsive predators are less able to maintain diversity. Additionally, I show that habitat partitioning does not interfere with the ability of distance-responsive predators to maintain diversity, even if it causes seedling survival to be highest near conspecific adults. From an empirical aspect, I studied the host range of seed-associated fungi. Soilborne microbes, such as fungi, are thought to play an important role in maintaining diversity in tropical forests. However, the microbial community itself is often treated as a black box, and little is known about which microbes are causing major effects, or how 8 specialized seed-microbe associations are. Here I use experimental inoculations to examine the host range and effect of a guild of seed-associated fungi that are thought to be mainly pathogens. I show that fungal species are differentially able to colonize different seed species, and have species-specific effects on seed germination. I show that in many cases, plant phylogeny, and to a lesser extent fungus phylogeny, are good predictors of colonization. Finally, I study how an optimally foraging granivore can promote (or undermine) coexistence amongst annual plants, using theory. Optimal foraging theory is one of the major theories for how predators behave; despite this, little is known about whether an optimally foraging predator could promote coexistence amongst a diverse community of prey. Previous models have shown than two species can coexist due to optimal foraging, but did not test whether multiple prey can coexist, nor if the effect is altered by environmental variation. Here, I show that if the predators specialize on different prey at different times, the predators can allow multiple prey species to coexist. In this case, environmental variation has little effect on the ability of predators to maintain diversity. If the predators are generalists, they cannot maintain diversity. Additionally, I show that generalist predators will create a negative storage effect, undermining coexistence.

Infection by multiple parasites is a part of everyday life for many organisms. The host immune system may be a central mediator of the many ways parasites might influence one another (and their hosts). Immunity provides a means for the colonized to reduce the success of current and future colonizers and has evolved across the tree of life several times independently. Along the way, the immune systems of plants as well as many groups of animals has evolved perhaps an accidental vulnerability wherein defense against one parasite can increase susceptibility to others. This so-called immune 'cross-talk' is a conundrum worth investigating not only to understand the impact of parasites on focal organisms, but also to better predict how immunity itself influences the evolution and epidemiology of parasites whose spread we might like to curtail. For plants, co-infection often comes from insect herbivores and various bacteria that colonize the leaf interior. Both colonizers can reduce plant fitness directly or indirectly by potentiating future enemies via cross-talk in plant immunity. This phenomenon has largely been studied in laboratory model plants, leaving a substantial gap in our knowledge from native species that interact in the wild. This dissertation helps close this gap by investigating the ecology of co-infection of a native plant by its major insect herbivore and diverse leaf-colonizing bacteria. I revealed that leaf co-infection in the field by leaf-mining herbivores and leaf-colonizing ("phyllosphere") bacteria is substantially more common than single infection by either group and that bacterial infection can cause increased feeding by herbivores in the laboratory. Immune cross-talk can also shape the field-scale patterns of herbivory across a native plant population. Studying the main herbivore of this native plant in detail revealed that, in contrast to many specialist herbivores, our focal species avoids plant defenses likely because it does not possess a specialized means of avoiding their toxicity. Nonetheless, this species may depend on the very same defenses it avoids by being initially attracted to plants that produce them. This foraging strategy is unique among known specialists. Lastly, I moved beyond immune cross-talk to explore how co-occurring phyllosphere bacteria might directly impact one another through competition. In the lab, I found that different growth strategies underlie competitive ability for two major clades of bacteria within the genus Pseudomonas, and that toxin production and resistance may be important mediators of competition within the phyllosphere. However, competitively superior bacteria that produce toxins may indirectly facilitate the survival of inferior competitors through their being toxin resistant, which likely enhances co-existence of diverse bacteria in the phyllosphere. Together, this dissertation has revealed a variety of means by which co-infecting bacteria and insects might influence one another through plant defense cross-talk, as well as how the complex interplay of colonization and competition might affect the structure of leaf microbial communities in nature.

Genome sequencing technologies are providing huge quantities of data for phylogenetic inference. However, most phylogenomic studies exclude gene families, because many have a complicated history of gene duplication/loss and structural change by domain shuffling, especially in deep phylogenies. Gene tree parsimony (GTP) methods, which seek the species tree that minimizes the cost of gene duplication, have been successfully applied to gene families with frequent duplication history. Their utility and performance in the context of gene families with complex histories of gene duplication and domain reshuffling remains unclear. In this study, we analyzed 4389 gene families from six angiosperm genomes encompassing a wide range of duplication rates, and a broad diversity of domain architecture. Overall species tree inference accuracy increased monotonically with the inclusion of more gene trees, and high accuracy was achieved with 50-100 gene trees. The rate of gene duplication strongly influences species tree inference accuracy, with the highest accuracy at either very low or very high rates of duplication and lowest accuracy centered around one duplication per branch in the unrooted species tree. This is the opposite of the relationship between substitution rates on tree construction accuracy, in which intermediate rates have highest accuracy. Accuracy is generally higher in gene families with high domain architecture diversity but has high variance in families with relatively low domain architecture diversity. The latter is probably due to the high variation of gene duplication number for those gene families. We close with some discussion of potential impacts of domain evolution on phylogenomic reconstruction protocols in general, including its effect on alignment.

Seed size is regarded as a functional trait with very important consequences for the fitness of plant species. Seedlings emerging from larger seeds are more competitive but are more costly to produce than seedlings from smaller seeds. Seed size is also a trait with transgenerational effects, affecting both the fitness of the parent as well as that of the offspring. Theory on the evolution of offspring size predicts an optimum balance between size and number, seen from the parent's perspective; while empirical studies often show selection for larger seeds, seen from the offspring's perspective. Seed size selection arising from post germination traits is, however, often not unidirectional, nor operating with the same strength in all life history stages of the plant. Seed size selection is also environmentally dependent. Even environmental influence might not operate with the same consistency and strength uniformly through the plant's life cycle. This dissertation is intended to study these questions concerning the dynamics of seed size selection in the wild. This work is to my knowledge, the first to document how seed size selection operates through the whole life cycle, with naturally germinated annual plants from the Sonoran Desert. In my first chapter I explored the offspring fitness consequences of seed size in a multiyear observational study using plant demography and relating vital rates (germination, survival, and fecundity) to the size of the seeds that originate individual plants and the environmental variables of precipitation and competition. I detected positive directional selection operating both through survival and fecundity. Water availability increased both survival and fecundity but also strengthened survival selection and had no effect on fecundity selection. Competition detrimental effects were only observed in fecundity but not in plant survival. In my second chapter I ask whether seed size-specific germination could influence seed size selection later in the life cycle. We found that because germination is differential in relation to seed size, the time of optimal conditions for germination in the field would determine the variance of seed size in the germinated fraction and thus influencing the strength of seed size selection operating through survival. In my third chapter I explored the dispersal consequences of phenotypic plasticity in seed provisioning. We found that mother plants that experienced more competition made smaller seeds and affected the seed dispersal process. Smaller seeds were better able to disperse farther away from their mothers and therefore increased their probability of escaping competition in the next growing season. These studies demonstrated that seed size selection varies through the life cycle and in intensity depending on interactions with the environment.

Mutualism is an often-complex positive interaction between species, each of which responds independently to varying biotic and abiotic conditions. Temperature is an important factor that can affect species both directly (e.g., physiologically) and indirectly (e.g., via its effects on interactions with consumers, competitors, and mutualists). Although much research has investigated the consequences of temperature for individual organisms, the effects of temperature on the formation, dissolution, and success of species interactions remain minimally understood. The unique ways in which species respond to temperature likely play a role in structuring communities. Environmental heterogeneity, including the thermal environment, can promote coexistence when species exploit resources in different ways, such as by occupying different thermal niches. This dissertation examines the consequences of temperature for participants in an ant-plant protection mutualism, and investigates how the thermal ecology of individual species affects the interaction. Many mutualisms involve multiple species, or interacting guilds. In these mutualisms, species interact with partner species that vary in multiple characteristics. Mutualists are quite sensitive to both partner quantity and partner quality (e.g., their effectiveness at performing a beneficial task). Mutualisms between ants and plants are common across a variety of habitats worldwide, which differ in thermal range, fluctuation, and seasonality. In light of ants’ well-studied and predictable responses to temperature, ant-plant interaction networks provide excellent systems for studying the thermal ecology of mutualisms. In ant-plant protection mutualisms, each of the participants (ants, plants, and enemies) likely differs in its response to temperature. In addition to the direct effects of temperature on ant species, temperature may affect the magnitude of mutualistic interactions among species by affecting the quantity and quality of the reward offered to partners, and the activity of the partners themselves and the plant’s enemies (i.e., herbivores). If herbivores are more thermally tolerant than the mutualistic ant defenders, the consequences for plants may well be severe; however, if herbivores are less thermally tolerant than are the ants, the effects of rising temperatures might be mitigated: although less-effective ants might be more frequent in a warmer world, herbivores would be less abundant there. This dissertation describes the thermal ecology of the participants in a mutualism between the cactus Ferocactus wislizeni and four of its common ant defenders (Forelius pruinosus, Crematogaster opuntiae, Solenopsis aurea, and Solenopsis xyloni) in the extreme environment of the Sonoran Desert, USA. The ants are attracted to extrafloral nectar produced by the plant, and in exchange protect the plants from herbivores, including a common phytophagous cactus bug, Narnia pallidicornis (Hemiptera: Coreidae). Specifically, it investigates how thermal ecology of the individual species affects the interactions among those species. Also, it considers the impact of a tradeoff between behavioral dominance and thermal tolerance among ants.

Dispersal influences the genetic and social composition of populations, yet it has been difficult to understand the mechanisms underlying dispersal and this limits our ability to understand how dispersal may be influencing population dynamics. Behavioral traits, such as aggression, have been implicated as drivers of both dispersal and population dynamics. However, the influence on both has never been addressed in a single system. Western bluebirds (Sialia mexicana) provide an excellent opportunity to address this question, as their dispersal propensity is dependent upon aggressive phenotype and we have detailed observations over a period of more than a decade. I show that natal dispersal is influenced by an interaction between father and son aggressive phenotypes, in addition to available resources on the natal territory. Furthermore, population density is influenced by resource availability and an interaction between population aggression and recruitment of offspring as breeders. Males that breed for multiple seasons once the population has reached saturation recruit a higher proportion of offspring into the population, as do males that are nonaggressive. Males that are nonaggressive are more likely to breed for multiple seasons, which suggests an added cost to aggressive behavior in this species. Both aggressive behavior and the availability of resources are mechanisms influencing dispersal of individuals that manifest at the population scale.

Increasingly, individual variation is being recognized as an important influence on behavioral evolution. Sources of variation are therefore an important target for research into the development, evolution, and function of behavior. By providing information about the timescale on which individuals are responsive to their environment, patterns of within-individual variation can shed light on function of behavioral variation. Here, I wanted to understand the function of behavioral variation and the genetic and environmental sources of variation in behavior. First, I test the hypotheses that variation in begging signals nestling hunger, need, or quality. Hunger is a short-term response to food deprivation, while need and quality give long-term information about fitness benefits of gaining more food and fitness potential, respectively. Second, I test the hypotheses that variation in begging is due to genetic, permanent environment, common environmental, and maternal effects. I test these hypotheses in the begging behavior of western bluebirds (Sialia mexicana), making repeated measurements across the nestling period. I show that begging behavior is consistent across the nestling period, and that nestling begging intensity increases with food deprivation. Nestlings fed during a given parental visit beg at higher intensity than nestmates, and on average wait longer since their last meal compared to individuals who were not fed in the same visit. These results support the hypothesis that variation in nestling begging signals hunger. I also show that responsiveness to food deprivation is negatively related to condition, but this effect is not consistent across the nestling period. Finally, variation in begging is produced by a common environmental effect that is correlated through time, suggesting that begging is strongly influenced by the nest environment. Together, these results indicate that variation in begging signals short-term changes in hunger and that environmental effects dominate the production of variation in begging.

The overarching goal of my dissertation is to explore the potential and limits of a trait-based approach to plant ecology. Together, the different studies presented here address two explicit and implicit foundational assumptions underpinning the trait-based approach: (1) that the correlation patterns and biological significance of traits transfer across scales and (2) that the phenotypic complexity of plants can accurately be synthesized into a few meaningful traits to study their ecology. Moreover, the last chapter focuses on a third key assumption: (3) that traits are strong predictors of plant performance (Shipley et al. In Press). I examine these assumptions by exploring multivariate patterns of phenotypic variation and integration across different ecological scales (e.g., individuals, populations, species) while explicitly considering the phenotypic complexity of trees, both in terms of their multidimensional and integrated nature. Two themes thus permeate this body of work: scales and phenotypic complexity. Much of what we know about the relationships among key traits comes from species-scale studies. Trait variation at smaller scales are often interpreted in the context of these interspecific relationships, but it is not clear that interspecific patterns observed at global scales apply to smaller scales. Moreover, although plants are complex, integrated organisms with intricate relationships among their traits, single traits are often studied and interpreted without considering the rest of the phenotype. Yet, examining individual traits outside of their phenotypic context might provide limited insight or be misleading. To address these shortcomings, this body of work examines multidimensional patterns of trait variation and correlation across ecological scales. It uses (1) a set of six ecophysiological leaf traits from mature trees in a lowland tropical rainforest, and (2) a set of twenty leaf, root, stem, branch and whole-plant ecophysiological traits from deciduous saplings in a temperate forest. The combination of our findings point to three main conclusions: (i) local interspecific and intra-population trait integration structures differ from each other and from the global interspecific patterns reported in the literature, such that global-scale interspecific patterns cannot readily be transferred to more local scales; (ii) considering the complexity of the plant phenotype provides better insights into ecological patterns and processes than what we can learn from considering individual or a handful of traits; and (iii) traits strongly affect individual plant performance, although there is no relationship between a species' trait correlation structure and its environmental niche, which suggests that there are multiple alternative optimal phenotypes in a given environment.

Amazonian evergreen forests are of broad interest, attributable to their ecological, economic, aesthetic, and cultural importance. However, their fate under climate change remains uncertain, largely due to unclear mechanisms in regulating tropical photosynthetic metabolism. Understanding mechanistic controls on these dynamics across time scales (e.g. hours to years) is essential and a prerequisite for realistically predicting tropical forest responses to inter-annual and longer-term climate variation and change. Tropical forest photosynthesis can be conceptualized as being driven by two interacting causes: variation due to changes in environmental drivers (e.g. solar radiation, diffuse light fraction, and vapor pressure deficit) interacting with model parameters that govern photosynthetic behavior, and variation in photosynthetic capacity (PC) due to changes in the parameters themselves. In this thesis, I aim to reveal photosynthetic controls by addressing three fundamental but complementary questions: (1) What are the mechanisms by which the subtle tropical phenology exert controls on tropical photosynthetic seasonality? (2) How do the extrinsic and intrinsic controls regulate the photosynthesis processes at hourly to interannual time scales in an Amazonian evergreen forest? (3) Are there sufficiently consistent relations among leaf traits, ages, and spectra that allow a single model predict the leaf aging process of Amazonian evergreen trees? To address question 1, I firstly show that seasonal change in ecosystem-scale photosynthetic capacity (PC), rather than environmental drivers, is the primary driver of seasonality of gross primary productivity (GPP) at four Amazonian evergreen forests spanning gradients in rainfall seasonality, forest composition, and flux seasonality. Using novel near-surface camera-detected leaf phenology to drive a simple leaf-cohort canopy model at two of these sites, I further show that leaf ontogeny and demography explain the changes in ecosystem photosynthetic capacity. The coordination of new leaf growth and old leaf divestment (litterfall) during the dry season shifts canopy composition towards younger leaves with higher photosynthetic capacity, driving large seasonal increases (~27%) in ecosystem photosynthetic capacity. To address question 2, I used the 7-year eddy covariance (EC) measurements in an Amazonian tropical evergreen forest. I used a statistical model to partition the variability of 7-year EC-derived GPP into two main causes: variation due to changes in extrinsic environmental drivers and variation in intrinsic PC. The fitted model well predicts variability in EC-derived GPP at hourly (R²=0.71) to interannaul (R²=0.81) timescales. Attributing model predictions to causal factors at different timescales, I find that ~92% of the variability in modeled hourly GPP could be attributed to environmental driver variability, and ~5% to variability in PC. When aggregating the modeled GPP into the annual time-step, the attribution is reversed (only ~4% to environment and ~91% to PC). These results challenge conventional approaches for modeling evergreen forests, which neglect intrinsic controls on PC and assume that the primary photosynthetic control at both long and short timescales is due to changes in the hourly-to-diurnal environment on the physiological phenotype. This work thus highlights the importance of accounting for differential regulation of different components of GPP at different timescales, and of identifying the underlying feedbacks and adaptive mechanisms which regulate them. To address question 3, I explored the potential for a general spectrally based leaf age model across tropical sites and within the vertical canopy profiles using a phenological dataset of 1831 leaves collected at two lowland Amazonian forests in Peru (12 species) and Brazil (11 species). This work shows that a simple model (parameterized using only Peruvian canopy leaves) successfully predicts ages of canopy leaves from both Peru (R²=0.83) and Brazil (R²=0.77), but ages for Brazilian understory leaves with significantly different growth environment and leaf trait values have lower prediction accuracy (R²=0.48). Prediction accuracy for all Brazilian samples is improved when information on growth environment and leaf traits were added into the model (5% R² increase; R²=0.69), or when leaves from the full range of trait values are used to parameterize the model (15% R² increase; R²=0.79). This work shows that fundamental ecophysiological rules constrain leaf traits and spectra to develop consistently across species and growth environment, providing a basis for a general model associating leaf age with spectra in tropical forests. In sum, in this thesis, I (1) conceptualize photosynthesis as being driven by two interacting dynamics, extrinsic and intrinsic, (2) propose and validate a model for biological mechanisms that mediate seasonal dynamics of tropical forest photosynthesis, (3) assess and quantify the factors controlling tropical forest photosynthesis on timescales from hourly to interannual, and (4) develop a general model for monitoring leaf aging processes of tropical trees across sites and growth environments. The revealed mechanisms (and proposed models) in this thesis greatly improve our mechanistic understanding of the photosynthetic and phenological processes in tropical evergreen forests. Strategic incorporation of these mechanisms will improve ecological, evolutionary and earth system theories describing tropical forests structure and function, allowing more accurate representation of forest dynamics and feedbacks to climate in earth system models.

Ecological systems are inherently dynamic, and a primary way in which they are dynamic is through time. Individual organisms, populations, communities, species interactions, and ecosystem functions all follow a temporal progression from the past, to the present, and into the future. This temporal progression can occur over the course of minutes, hours, days, weeks, months, years, decades, or various other timescales. In this sense, temporal dynamics are an intrinsic property of all biological systems. In fact, one of the most prominent signals of recent global climate change is the significant change in the timing of biological events for a diversity of organisms. In light of this widespread pattern, there is a renewed interest in understanding the multifaceted importance of time in ecology. In this dissertation, I investigate the temporal ecology of a subalpine ecosystem, specifically focusing on flowering plant communities and plant-pollinator interactions. I examine the temporal dynamics of this system over multiple decades in response to ongoing climate change as well as over shorter time scales within a growing season. Using a 39-year record of flowering phenology, I show that species-specific shifts in the timing of flowering in response to climate change can substantially reshape a subalpine plant community over this time period. Community phylogenetic analyses reveal that these changes are largely independent of evolutionary history. Using a laboratory experiment, I show that the timing of an important harsh abiotic event-low temperatures that cause frost damage to plants-can differentially affect flowering plant species, with implications for plant demography, community structure, and interactions with pollinators. Finally, I show that plant-pollinator interactions exhibit substantial within-season temporal turnover, and that this temporal flexibility of plant-pollinator interactions from one week to the next is consistent and predictable across years. Taken together, this dissertation provides a multifaceted investigation of the temporal ecology of plant communities and plant-pollinator interactions, revealing the important consequences of ecological timing at short-term and longer-term scales.