Ants, stem-borers, and fungal pathogens: experimental tests of a fitness advantage in Piper ant-plants

This study tests experimentally the hypothesis that Pheidole bicornis ants increase the fitness of Piper ant-plants (Piperaceae) in Costa Rican forests. In two experiments with naturally occurring Piper ant-plants, ∼ 50 individuals were randomly selected and maintained either as controls (with ants) or as ant-exclusion plants (without ants) for 2 yr. Leaf replacement rates and seed production, as measures of relative plant vigor and reproductive potential, were significantly greater in control plants with intact ant colonies than in plants from which ant colonies were experimentally excluded by periodical treatment with dilute insecticide. A series of experiments was designed to assess the relative contribution of antiherbivore defense, nutrient procurement, and antipathogen defense as potential mutualistic functions of the ants. To assess the fitness effects of antiherbivore defense, leaf replacement rates and seed production were measured on plants to which artificial folivory was applied over a 2-yr period (33% of each leaf blade removed), and stem-boring weevil damage was monitored on plants with and without ants. Nutrient procurement by ants was estimated quantitatively, and net leaf production was monitored experimentally on control plants (with ants), on plants without ants, and on nutrient-enriched plants with ants (microquantities of fertilizer added to lower stem). Epiphyll loads, phylloplane spore densities, and disease incidence were compared on experimental plants with and without ant colonies. The results indicate that folivory, the classical parameter measured in ant-plant studies, was not related to differences in fitness in Piper ant-plants; instead, ant disruption of stem-borers and ant foraging on inflorescences to reduce fungal invasion were identified as probable mechanisms by which ants conferred an average fitness advantage of at least 60 and 10%, respectively, in the 2nd yr of the comparison. Although neither antifolivore defense nor nutrient provisioning by ants appeared to benefit Piper plants, they may contribute synergistically or may operate on temporal or spatial scales not included in the study. For example, nutrient provisioning may allow Piper ant-plant species to occupy and compete favorably in poor-soil habitats not evaluated in these experiments.