ABSTRACT: Traits are important for understanding how plant communities assemble and function, providing a common currency for studying ecological processes across species, locations, and habitat types. However, the majority of studies relating species traits to community assembly rely upon vegetative traits of mature plants. Seed traits, which are understudied relative to whole-plant traits, are key to understanding assembly of plant communities. This is particularly true for restored communities, which are typically started de novo from seed, making seed germination a critical first step in community assembly and an early filter for plant establishment. We experimentally tested the effects of seed traits (mass, shape, and embryo to seed size ratio) and phylogeny on germination response in 32 species commonly used in prairie grassland restoration in the Midwestern USA, analyzing data using time-to-event (survival) analysis. As germination is also influenced by seed dormancy, and dormancy break treatments are commonly employed in restoration, we also tested the effects of two pretreatments (cold stratification and gibberellic acid application) on time to germination. Seed traits, phylogeny, and seed pretreatments all affected time to germination. Of all traits tested, variables related to seed shape (height and shape variance) best predicted germination response, with high-variance (i.e., pointier and narrower) seeds germinating faster. Phylogenetic position (the location of species on the phylogenetic tree relative to other tested species) was also an important predictor of germination response, that is, closely related species showed similar patterns in time to germination. This was true despite the fact that all measured seed traits showed phylogenetic signal, therefore phylogeny provided residual information that was not already captured by measured seed traits. Seed traits, phylogenetic position, and germination pretreatments were important predictors of germination response for a suite of species commonly used in grassland restoration. Shape traits were especially important, while mass, often the only seed trait used in studies of community assembly, was not a strong predictor of germination timing. These findings illustrate the ecological importance of seed traits that are rarely incorporated into functional studies of plant communities. This information can also be used to advance restoration practice by guiding restoration planning and seed mix design.
