For insect-pollinated plants, pollen dispersal strongly depends on the movement behavior of pollinators. Unlike many of the other factors that contribute to successful pollination (e.g., pollen compatibility), the movement step is highly sensitive to ecological context, and may vary significantly depending on the spatial arrangement and background community of the donor and recipient plants. The objective of this study was to examine how this ecological context affects pollinator visitation behavior using constructed arrays of plants in the field. We used two species of thistle, Carduus nutans and C. acanthoides, in pure and mixed arrays of different densities, and recorded pollinator movements within each array.

Results/Conclusions

Preliminary results suggest that pollinator movements differ depending on the species composition of the array. Self-pollinating moves were less likely to occur than non-self moves in pure C. nutans arrays (p=0.07). In addition, pollinators moved longer distances between plants in pure C. nutans arrays (p=0.03). C. nutans produces larger, but fewer, flower heads than C. acanthoides, and thus may be more attractive to foraging pollinators. Pollinators may be willing to move farther if the destination inflorescence is sufficiently rewarding given the energy expenditure. In mixed species arrays, pollinator moves to C. nutans occurred more frequently than expected by chance (i.e., out of proportion with C. nutans abundance, p=0.008), whereas pollinator moves to C. acanthoides did not occur out of proportion with C. acanthoides abundance (p=0.31). Pollinators appear to have a stronger preference for C. nutans compared to C. acanthoides.

We have previously developed a mechanistic gravity model of pollinator visitation behavior, in which the probability of pollinator movement between inflorescences is determined by the properties of these two inflorescences, mediated by the distance between them. We conclude that the results from our study are consistent with this gravity model. Pollinator movement behavior changed depending on the context of the array. Understanding the mechanism by which ecological context produces directionally-biased movement of pollinators may lead to important insights on patterns of gene flow via pollen.