The effect of pollen source on seed traits and dispersability in the heterocarpic annual Crepis sancta

Aims: The evolution and expression of dispersal-related traits are intertwined with those of other life-history functions and are manifested within various physiological constraints. Such a relationship is predicted between inbreeding levels and dispersability, which may be anatomically and ontogenetically linked so that the selection pressures on one may affect the other. While both the effect of inbreeding on reproductive success and on dispersal strategies received much attention, only a few studies considered both simultaneously. Furthermore, such studies often rely on two dichotomic representations of breeding and dispersal: Using selfing versus outcrossing as a representation of breeding level, and dispersal ratio as the sole representation of dispersal strategy. Methods: Here, we used pollination experiments in the heterocarpic Crepis sancta (Asteraceae) to expand in two different manners on the common practice of using dichotomic representations of breeding and dispersal. First, we used pollination treatments that represent a continuum from selfing through pollination by kin to pollination by a distant neighbor. Second, we measured a whole set of continuous morphological and dispersal-related traits, in addition to measurements of reproductive success and dispersal ratio. Important Findings: The proportion of developed capitula and the number of both dispersed and non-dispersed achenes were significantly lower in the self-pollination treatment in comparison to the outcrossed treatments. The effect of pollen sources on dispersal ratio was not statistically significant, though self-pollinated plants rarely produced non-dispersing seeds. Achene's biomass increased with distance between parent plants, but pappus width did not, leading to a nonsignificant effect of pollination on falling velocity. Overall, pollen source affected mainly traits that were associated with reproductive output, but it had no clear effect on predominately dispersal-related traits. Such differences in the response of reproduction and dispersal traits to variation in pollen source suggest that dispersal-related selection is probably weak and/or masked by other forces.