Real-time evolution supports a unique trajectory for generalized pollination

Florian P Schiestl,Alice Balmer,Daniel D Gervasi
DOI: https://doi.org/10.1111/evo.13611
IF: 4.171
Evolution
Abstract:Whereas specialized pollination is well recognized to cause floral adaptation, we know little about the evolutionary impact of generalized pollination. For example, it is largely unknown whether such pollination can lead to adaptive floral divergence and to what degree pollinators with different effectiveness determine evolutionary trajectories. Here, we investigated the evolutionary consequences of combined bumblebee- and hoverfly-pollination ("generalized" pollination) in comparison with those of each individual pollinator species (specialized pollination), using fast-cycling Brassica rapa plants during seven generations of experimental evolution. Bumblebees were twice as efficient as hoverflies in pollinating B. rapa flowers, but phenotypic selection and evolutionary change in plants with generalized pollination was different from both bumblebee- and hoverfly-pollinated plants for several traits. After seven generations evolution, plants with generalized pollination resembled bumblebee-pollinated plants in having little spontaneous selfing and tall size, but were more similar to hoverfly-pollinated plants in having low floral scent emission. This unique trait combination supports the idea of a generalized-pollination ecotype, coined neither by the most efficient pollinator, nor by an evolutionary average between the changes caused by each individual pollinator. For a better understanding of such "nonadditive evolution," future research should target interactions of pollinators and their effect on phenotypic selection.
What problem does this paper attempt to address?