Pollination and herbivory treatments
For the pollination treatments, all flowers on a plant were either exposed to ambient pollination conditions (control) or received supplemental hand pollinations every ca. 5 days for the full flowering period. Only three plant genotypes were used as pollen donors to control for pollen quality effects. Herbivory was manipulated through the addition or removal of an ecologically important herbivore; the strawberry leaf beetle, Galerucella tenella L. (Coleoptera: Chrysomelidae). This oligophagous herbivore feeds on leaves and flowers of several Rosaceae plants (Stenberg & Axelsson 2008). For woodland strawberry this can reduce fitness both directly and indirectly, via pollinator limitation (Muola et al. 2017; Muola & Stenberg 2018). The plantation in which the herbivore treatments were applied was fenced with a 2 m fine mesh to exclude herbivory from small and large browsing vertebrates, including deer and digging rodents. For the ‘herbivore addition’ treatment, several hundred adult individuals of the strawberry leaf beetle were collected in early May soon after their emergence on their main host plant meadowsweet (Filipendula ulmaria (L.) Maxim.) (Stenberg & Axelsson 2008). Further details of these collection localities are provided in (Weber et al. 2020b). The beetles were released onto plants at a density of 0.4 individuals per plant. Adult and larval feeding damage were apparent within one and five weeks of release respectively. Most feeding damage in the ‘herbivore addition’ treatment was hence caused by the strawberry leaf beetle. However, damage from larvae of two species of leaf‐ and flower-feeding Lepidoptera – Cnephasia asseclana Denis & Schiffermüller (Tortricidae) and Ceramica pisi L. (Noctuidae) – was also observed on experimental plants. The former is a pest of cultivated strawberry (Sigsgaard et al. 2014), whereas the latter is a generalist moth that feeds on genera including Rubus andSalix (Robinson et al. 2010). As a common practice for the experimental removal of insect herbivores (Siemann et al. 2008), the ‘herbivore removal’ treatment was made through application of low doses of an insecticide. For this we employed foliar applications of Calypso (Bayer CropScience); a systemic insecticide based on the active substance thiacloprid. All manufacturers’ recommendations were followed concerning dosage concentration and frequency of application. Three rounds of application were made every ca. 4 weeks between early May and mid-July. Spraying took place when conditions were dry and windless so as to avoid cross-contamination. Virtually no damage to plants in the ‘herbivore removal’ treatment was observed.