Summary1. Ecological restoration is a global priority that holds great potential for benefiting natural ecosystems, but restoration outcomes are notoriously unpredictable. Resolving this unpredictability represents a major, but critical challenge to the science of restoration ecology. 2. In an effort to move restoration ecology toward a more predictive science, we consider the key issue of variability. Typically, restoration outcomes vary relative to goals (i.e. reference or desired future conditions) and with respect to the outcomes of other restoration efforts. The field of restoration ecology has largely considered only this first type of variation, often focusing on an oversimplified success vs. failure dichotomy. The causes of variation, particularly among restoration efforts, remain poorly understood for most systems. 3. Variation associated with restoration outcomes is a consequence of how, where and when restoration is conducted; variation is also influenced by how the outcome of restoration is measured. We propose that variation should decrease with the number of factors constraining restoration and increase with the specificity of the goal. When factors (e.g. harsh environmental conditions, limited species reintroductions) preclude most species, little variation will exist among restorations, particularly when goals are associated with metrics such as physical structure, where species may be broadly interchangeable. Conversely, when few constraints to species membership exist, substantial variation may result and this will be most pronounced when restoration is assessed by metrics such as taxonomic composition. 4. Synthesis and applications. The variability we observe during restoration results from both restoration context (how, where and when restoration is conducted) and how we evaluate restoration outcomes. To advance the predictive capacity of restoration, we outline a research agenda that considers metrics of restoration outcomes, the drivers of variation among existing restoration efforts, experiments to quantify and understand variation in restoration outcomes, and the development of models to organise, interpret and forecast restoration outcomes.

Summary 1.Rainfall is a key determinant of production and composition in arid and semi-arid systems. Longterm studies relating composition and water availability primarily focus on current-year precipitation patterns, though mounting evidence highlights the importance of previous-year rainfall particularly in grasslands dominated by perennial species. The extent to which lagged precipitation effects occur in annual grasslands, however, remains largely unexplored. 2. We pair a long-term study with two manipulative experiments to identify patterns and mechanisms of lagged precipitation effects in annual grasslands. The long-term study captured variation in functional group (exotic annual forbs and grasses) abundance and precipitation across 8 years at three northern California grassland sites. We then tested whether lagged rainfall effects were created through seed production and litter (residual dry matter, RDM) by manipulating rainfall and litter, respectively. 3. Rainfall from the previous-year growing season (both seasonal and total rainfall) shifted functional group abundance. High lagged rainfall was associated with increased grass and decreased forb abundance the following year. Current-year seasonal rainfall also influenced species composition, with winter rain increasing forb and decreasing grass abundance. Lagged precipitation effects were generally stronger for forbs than for grasses. Our experimental studies provided evidence for two mechanisms that contributed to lagged effects in annual grasslands. Higher rainfall increased seed production for grasses, which translated to more germinable seed the following year. Higher rainfall also increased biomass production and RDM, which benefited grasses and reduced forb abundance. 4. Synthesis. Our results highlight the importance of previous-year precipitation in structuring annual community composition and suggest two important biotic pathways, seed rain and RDM, that regulate lagged community responses to rainfall. Incorporating lagged effects into models of grassland diversity and productivity could improve predictions of climate change impacts in annual grasslands.

Summary1. Post-dispersal seed predators contribute substantially to seed loss across many ecosystems. Most research has focused on understanding sources of variation in seed loss, without appreciating the implications of seed predation for plant coexistence, community assembly and broader community theory. Meanwhile, research aimed at understanding coexistence and community assembly processes in plant communities has focused on axes of dispersal and resource competition and the traits influencing these processes, without accounting for the role of generalist seed predators. 2. We review the unique features of post-dispersal seed predation and assess the implications of seed loss on three critical components of plant community organization -coexistence, community structure and plant invasions -pointing to both important gaps in theory and empirical knowledge. We highlight how understanding fundamental controls on plant recruitment is central to determining how seed predation affects plant recruitment and coexistence. We discuss how accounting for seed predator foraging strategies may shift trait-based inferences of community assembly. 3. Synthesis. We argue that seed predation by generalist consumers, which is pervasive in temperate communities, should be better incorporated into plant community theory. Experiments that specifically incorporate the presence and attributes of the seed predator community and that follow seed fate would fill important knowledge gaps. Particularly needed are studies focused on strengthening the connections between seed removal and plant establishment and linking selective and density-dependent foraging strategies to plant traits. Advancing our understanding of the processes regulating plant coexistence and community assembly requires that future research not only acknowledge but also incorporate generalist consumers' effects on plant communities.