Disruptive selection via pollinators and seed predators on the height of flowers on a wind‐dispersed alpine herb

• Published in: American journal of botany 2022-11, Vol.109 (11), p.1717-1729
• Main Authors: Chen, Kai‐Hsiu, Pannell, John R.
• Format: Article
• Language: English
• Subjects: andromonoecy; Animal reproduction; Animals; Antagonists; Breeding success; Dispersal; Dispersion; Females; Fertilization; floral display; floral evolution; Flowers; Flowers - anatomy & histology; Fruit; Grasslands; Growing season; Herbivory; Herbs; Insecta; Insects; Leaves; Maturation; Phytomyza; Plants; Plants (botany); Pollination; Pollinators; Positive selection; Predation; Predators; Ranunculaceae; Reproduction; Seed dispersal; Seed predation; Seeds; selection differential; selection gradient; Sex; sex allocation; Special Issue; Success; Vegetation; Wind
• ISSN: 0002-9122; 1537-2197
• DOI: 10.1002/ajb2.16073

Premise Floral stalk height is known to affect seed dispersal of wind‐dispersed grassland species, but it may also affect the attractiveness of flowers and fruits of animal‐pollinated and animal‐dispersed plants. Stalk height may thus be responsive to selection via interactions with both mutualist pollinators and seed dispersers, but also antagonist florivores and seed predators. In this study, we aimed to determine the effect of pollinators and seed predators on selection on floral stalk height in the insect‐pollinated and wind‐dispersed, alpine, andromonoecious herb Pulsatilla alpina, whose flowers also vary in their sex allocation and thus in the resources available to both mutualists and antagonists. Methods We measured the resource status of individuals in terms of their size and the height of the vegetation surrounding plants of P. alpina at 11 sites. In one population, we recorded floral stalk height over an entire growing season and investigated its association with floral morphology and floral sex allocation (pistil and stamen number) and used leaf‐removal manipulations to assess the effect of herbivory on floral stalk height. Finally, in four populations, we quantified phenotypic selection on floral stalk height in four female components of reproductive success before seed dispersal. Results Stalk height was positively associated with female allocation of the respective flower, the resource status of the individual, and the height of the surrounding vegetation, and negatively affected by leaf removal. Our results point to disruptive selection on stalk height in terms of both selection differentials and selection gradients for fertilization, seed predation, and seed maturation rates and to positive selection on stalk height in terms of a selection differential for mature seed number. Conclusions Stalk height of P. alpina is a costly trait that affects female reproductive success via interactions with both mutualists and antagonists. We discuss the interplay between the resource status and selection imposed on female reproductive success and its likely role in the evolution of sex‐allocation strategies, especially andromonoecy.