Plant sex effects on biotic interactions in dioecious willow

Sammanfattning: Sex in dioecious plants affects interactions between plant individuals and the organisms consuming them. Plant-herbivore interactions are affected by plant sex across many plant families, and plant-pathogen infections can be strongly sex-biased. Non-antagonistic consumers of plant materials can also be affected by sex; pollinators often prefer plants of either plant sex. Still, much remains to be investigated of the mechanisms underlying these plant sex effects. In addition, most knowledge of plant-sex biased interactions focus on plant individuals and much less is known of how these effects scale up to groups of plants or populations. Finally, applications of plant sex effects in agricultural systems are lacking. I established a large-scale field experiment using plots of a dioecious bioenergy crop, the willow Salix viminalis, to investigate what effects individual plant sex, and groups of males, females or sex-mixes, have on consumer organisms. I measured abundances of the most serious insect pest (Phratora vulgatissima), and one of its main predators (Anthocoris nemorum). Furthermore, I measured damage caused by roe deer (Capreolus capreolus) and infection of the most serious pathogens (Melampsora spp.) affecting S. viminalis. Finally, I measured effects of S. viminalis sex on pollinator visitation and on pollination in nearby strawberry. I performed laboratory-and greenhouse based controlled experiments to investigate mechanisms underlying the consumer interactions. I found that female plants were more seriously affected by the pathogens but that infections were unaffected by plant sex ratio of plots. Neither the pest nor its predator was affected by plant sex in the field experiments despite effects on some life history traits in laboratory experiments. Roe deer browsing increased in mixed-sex plots, but roe deer did not discriminate between male and female individual plants. Pollinators more often visited male plants but mixed-sex S. viminalis still increased nearby strawberry fruit production. Conclusions of this thesis are that plant-consumer species pair specificity matter for plant sex effects and that effects from individual plant interactions may differ from effects on interactions between plant groups and consumers. A better understanding of underlying mechanisms may be key to understanding how plant sex effects can be predicted and generalized. I also conclude that it is likely better to use male, than female, genotypes for S. viminalis plantations.