Are edges and young forests always preferred by leaf-cutting ants? The case of Acromyrmex in the Atlantic forest

QUEIROZ, JARBAS M; ELIZALDE, LUCIANA

Encuentro; Annual Meeting of the Association of Tropical Biology.; 2012

Leaf-cutting ants (Atta and Acromyrmex) are dominant herbivores in neotropical forests. Previous studies have shown that nest density of Atta species is higher in young secondary forests and forest edges, but information about the response of Acromyrmex species is scarce. In this study we asked whether Acromyrmex nest density at different distances from forest edges would show the same pattern of Atta species. We also compared Acromyrmex nest density between small or big forest fragments (i.e., with more or less border effect, respectively) and regeneration state (young or old secondary forest areas). Twelve forest fragments in the Parayba river basin and one forest remnant in Guapiaçu river basin, both located in Rio de Janeiro State, were sampled for ant nests and trails, with 10m width-transects going from edge to the most central part of forest fragments or remnant. We found seven Acromyrmex species, including the three subspecies of A. subterraneus. The most abundant species was A. niger that represented 63% of all nests sampled. Acromyrmex density did not differ according to fragment size (12.4 vs 9.6 nests per ha in big or small fragments, respectively, Wilcoxon paired test, P = 0.83), neither we found a border effect (Kendall correlation, P = 0.89). Nest density was five times higher in old than in young forest. Results were qualitatively the same using the most abundant species only. Acromyrmex and Atta seem to respond in different ways to forest edges and age, and this could be an important difference when considering the impact that leaf-cutting ants, as a group, exert over their environment. From the three current hypotheses to account for leaf-cutting ant nest distribution, our results show that the suitable soil hypothesis seems to be the more general one, as it may applies to both genera.