Background/Question/Methods

Tropical plant-ant interactions represent important model systems for ecological studies. In this study we examined what determines the distribution of Piper immutatum Trel. (Piperaceae) and its mutualistic ant species: Pheidole sp nov. (Myrmicinae). We observed how the interactions between these plants and its ants vary across altitude and tested the hypothesis that ants enhance plant survival. Our approach combined observational and experimental data and included a broad altitudinal range of distribution of P. immutatum and their ants. We evaluated the effects of mutualistic ants on herbivory and plant survival rates across a range of environmental conditions. We transplanted P. immutatum and excluded ants from the plants in a factorial design where we excluded the ants for a year at 12 sites, four of which where of which where outside of the plant’s natural range (400-2600m). We used MANOVA and profile analysis to examine the effects of ant exclusion on specialist and generalist herbivory and plant survival at lowland versus montane forests.  Lastly we utilized path analysis to test the causal hypothesis that mutualistic ants enhance plant survival outside the plant’s distribution.

 Results/Conclusions            Mutualistic ants enhanced the establishment of P. immutatum in montane forests, while it was costly to sustain ant colonies at lowland rain forests. The plant is most abundant at the 400-1600m range, and within this range, leaf production and herbivory rates do not covary with elevation.  After a year of ant exclusion we found that ant presence significantly enhanced the survival of P. immutatum in montane forests.  Leaf area gained was higher for plants without ants at lowland forests; presumably plants without ants were able to allocate more energy to leaf growth. Ant exclusion enhanced generalist herbivores in lowland forests but it significantly decreased generalist herbivory in montane forests. In contrast, ant exclusion decreased specialist herbivory in lowland forests and increased specialist herbivory in montane forests. Our combined results demonstrate that P. immutatum-Pheidole mutualism is strongest in montane forests, where ants reduce specialist herbivory, resulting in increased survival. Finally we found survival rates of plants outside their range of distribution where 2 fold higher when plants had ant colonies. Thus, the distribution of this plant may be restrained by the effects of climate on the mutualistic ant more so than on direct climatic effects on the plant.