Indirect effects of two types of insect herbivores on flower visitor community and their implications for plants’ reproductive performances

Background/Question/Methods

Since a wide range of arthropods visit flowers in seeking for resources, arthropod communities are formed on flowering plants. Decisions of flower visits of arthropods are influenced not only by plant traits but also by other interacting arthropods on plants. Recent studies have clarified herbivory during the vegetative season indirectly influenced floral traits, such as size, nectar and pollen in quality and quantity, thereby altering frequency of pollinator visits. No studies, to date, have explored consequences of herbivore-induced effects on whole community members on flowers: pollinators, florivores, larcenists, and predators. Non-pollinators may negatively influence plants’ reproductive performances by directly consuming flowers or indirectly decreasing in floral attractiveness to pollinators. Using tall goldenrods and its dominant herbivores, we ask whether herbivory in the vegetative season influences flower visitor community and plant reproduction through phenotypic changes in flowers. Specifically, we used two dominant herbivores: lacebugs and aphids with inducing different growth pattern of goldenrods, which may indirectly influence flower visitors and seed production via changes in floral traits. We set inoculation experiments using lacebugs and aphids to assess their herbivory effects on floral traits, flower visitors, and seed production. Additionally, we conducted supplemental hand pollination to evaluate the extent of pollen limitation.

Results/Conclusions

Lacebug-inoculated plants had lower flower numbers and amount of nectar, and smaller floral size than non-inoculated plants. Correspondingly, frequencies of flower visitors were lower in inoculated plants. In contrast, aphid-inoculated plants had greater flower numbers and similar amount of nectar and floral size, than non-inoculated plants, resulting in an increase in some flower visitor groups. Irrespective of changes in floral traits and visitor composition, the seed production was similar among treatments. Supplemental pollination did not increase seed:ovule ratio, suggesting little pollen limitation. In conclusion, herbivory during the vegetative season has a greater impact on annual plants’ reproductive output via changes in flower production (i.e., ovule production) depending on herbivore species identity, rather than via changes in flower visitor frequencies (i.e. fertilization success).