Factors related to the inter-annual variation in plants' pollination generalization levels within a community

The number of pollinators of a plant species is considered a measure of its ecological generalization and may have important evolutionary and ecological implications. Many pollination studies report inter-annual fluctuations in the composition of pollinators to particular species. However, the factors causing such variation are still poorly understood. Here we investigate how flowering duration and plant and pollinator assemblages influenced the inter-annual changes in the functional generalization level of the 20 most common plant species of a semi-natural meadow in southern Norway. We also studied the extent to which changes in generalization levels were controlled by flower-shape and flowering time. Large inter-annual changes in generalization levels were common and there was no relationship between the generalization level one year and the following. Generalization level of particular plant species increased with flowering duration, sampling effort, and the abundance of managed honeybees in the community. Generalization level decreased with the flowering synchrony between the focal plant species and the rest of the plant community and with the focal species' own abundance, which we attribute to inter-specific competition for pollinator attraction and foraging decisions made by pollinators. Plants with different flower-shapes and flowering times did not differ in the extent of inter-annual variation in generalization levels. Most studies do not consider the effect of the plant community on the generalization level of particular plant species. We show here that both pollinator and plant assemblages can affect the inter-annual variation in generalization levels of plant species. Studies like ours will help to understand how pollination interactions are structured at the community level, and the ecological and evolutionary consequences that these inter-annual changes in generalization levels may have.