Pollen limitation is widespread in plant populations. This phenomenon is most often explained as the result of the adaptive response of plants to variable pollen receipt, which encourages production of ovules in excess of the average number pollinated. The activity of pollination mutualism exploiters, e.g. nectar robbers, may also influence plant traits related to pollination and reproduction, but such effects have received less attention.
We investigated a population of Odontonema cuspidatum – a hummingbird-pollinated species with a prolonged bloom period – experiencing high levels of nectar robbery from stingless bees, to ask whether a trade-off between pollinator attraction and avoidance of nectar robbers can explain the extent of pollen limitation exhibited by the population. Using field surveys and experimental arrays of plants grown in a common garden, we asked 1) what effect floral traits have on pollination, robbery, and reproductive output and 2) whether floral traits, particularly flower number, show evidence of adaptive response to a trade-off between pollinator attraction and nectar robber avoidance. We pair empirical data with a conceptual model of reciprocal responses between floral display on the one hand and plant mutualists and antagonists on the other.
Results/Conclusions
cuspidatum shows strong pollen limitation, with pollen supplementation substantially increasing both fruit set and seeds/fruit. At the same time, nectar robbery reduces probability of a flower setting fruit by >40%. Pollinator visitation is positively related to floral display size, but intensity of nectar robbery also increases with floral display size, and at a faster rate than pollinator visitation. This suggests conflicting selection on floral display size in O. cuspidatum. A conceptual model indicates that balancing these forces would result in floral display sizes that are consistent with empirical observations but smaller than those that would be favored in the absence of nectar robbery.
These results suggest that strong pollen limitation in O. cuspidatum may in part result from plants’ adaptive response to conflicting pressure from mutualists and floral antagonists. An ongoing common garden experiment in which both pollination and nectar robbery are manipulated – which will be completed in June 2018 – will further test whether plants have plastic responses to pollination and/or nectar robbery intensity that allow them to adjust floral display size to maximize reproductive success.