Plants interact extensively with their neighbors, but the evolutionary consequences of variation in neighbor identity are not well understood. Seedling traits are likely to experience selection that depends on the identity of neighbors because they influence competitive outcomes. To explore this, we evaluated selection on seed mass and emergence time in two California grasses, the native perennial Stipa pulchra and the non-native annual Bromus diandrus, in the field with six other native and non-native neighbor grasses in individual and mixed species treatments. We also quantified characteristics of each neighbor treatment to further investigate factors influencing their effects on fitness and phenotypic selection. Selection favored larger seeds in both focal species and this was largely independent of neighbor identity. Selection generally favored earlier emergence in both focal species, but neighbor identity influenced the strength and direction of selection on emergence time in S. pulchra but not B. diandrus. Greater light interception, higher soil moisture, and greater productivity of neighbors was associated with more intense selection for earlier emergence and larger seeds. Our findings suggest that changes in plant community composition can alter patterns of selection in seedling traits, and that these effects can be associated with measurable characteristics of the community.

Data were collected in two growing seasons in 2017-2018 from a field experiment at the University of California San Diego Biological Field Station (California, USA).

The Data folder contains 7 separate datasets as CSV files, each with accompanying .txt metadata files:
1) A dataset of individual-level data (Waterton2023_SeedlingSelection_Individual.csv).
2) A dataset of individual-level seed production data for 2017 (Waterton2023_SeedlingSelection_Seed_2017.csv).
3) A dataset of individual-level seed production data for 2018 (Waterton2023_SeedlingSelection_Seed_2018.csv).
4) A dataset of neighbor emergence (Waterton2023_SeedlingSelection_Neighbor_Emergence.csv).
5) A dataset of aboveground biomass in neighbor treatments (Waterton2023_SeedlingSelection_Neighbor_Biomass.csv).
6) A dataset of photosynthetically active radiation (PAR) in neighbor treatments (Waterton2023_SeedlingSelection_Neighbor_PAR.csv).
7) A dataset of volumetric water content (VWC) in neighbor treatments (Waterton2023_SeedlingSelection_Neighbor_VWC.csv).

R statistical software is required to run the R script.