Fruit secondary metabolites decrease the quantity and quality of secondary seed dispersal

Background/Question/Methods

Because plants are rooted in place, they often rely on animals to disperse their seeds. While the primary function of fleshy fruits is to attract seed-dispersing animals, fruits often contain complex mixtures of secondary metabolites that can be toxic or deterrent to consumers. These compounds may function as defense against seed predators and pathogens, but this can come at the cost of also reducing the quantity or quality of seed dispersal. Past studies have demonstrated that secondary metabolites in fruits can indeed result in trade-offs between defense and dispersal, but because seeds are often dispersed across multiple stages by both primary and secondary dispersers, little is known about the ultimate costs of secondary metabolites for plant movement. Here, we tested the effects of amides, a class of nitrogen-based defensive compounds common in fruits in the neotropical plant genus Piper (Piperaceae), on the effectiveness of secondary seed dispersal by ants. Specifically, we asked: (1) Do ants serve as effective secondary seed dispersers of Piper reticulatum?, (2) Do amides in fruits reduce the quantity or quality of secondary seed dispersal?, and (3) What might the net effects of amides be on seed dispersal success and plant movement?

Results/Conclusions

We found that ants serve as secondary seed dispersers of P. reticulatum, strongly recruiting to and removing entire ripe and overripe fruits (but not unripe fruits or cleaned seeds) on the forest floor. By experimentally adding amide extracts to Piper fruits in the field, we found that amides reduced the quantity of secondary seed dispersal by reducing ant recruitment (87%) and fruit removal rate (58%). Interestingly, although amides did not affect the distance ants initially carried seeds, amides altered seed dispersal quality by reducing the likelihood of ants cleaning seeds (67%) and increasing the likelihood of ants redispersing seeds outside of the nest (200%). Overall, these findings suggest that amides affect not only the likelihood of secondary seed dispersal but also the spatial distribution of dispersed seeds (i.e., whether seeds are deposited inside or outside of the ant nest) and ant movement in the landscape. Overall, these results demonstrate that fruit secondary metabolites can alter the effectiveness of secondary seed dispersal through multiple mechanisms, and understanding the consequences for plant fitness and demography will require evaluating the effects of seed handling and deposition location on germination.