Much of the ecological alteration faced by human-modified Neotropical forests can be assigned to edge effects, including the proliferation of some voracious herbivores such as leaf-cutting ants. However, the underlying mechanisms/impacts of tropical forest edge on herbivores performance and their foraging behaviour (e.g. dietary diversity) have rarely been investigated. The goal of this study was, therefore, to determine whether and how the annual diet (i.e. species richness, diversity and the relative proportion of pioneer versus non-pioneer species of plant materials) of Atta cephalotes colonies differs in the forest edge versus the interior zone of a large remnant of Atlantic forest in northeastern Brazil. Among the key results was a strong habitat effect on dietary diversity (explaining ca. 40–50% of the variation), which, in edge colonies, decreased approximately by one fourth compared to interior colonies (inverse of Simpson's index: 3.7±0.84 versus 4.99±0.95). There was a predominance of leaf fragments collected from pioneer species in the diet in both habitat (86% in edge and 80.4% in interior). Edge colonies collected proportionally more fragments from pioneer species than colonies located in the forest interior. Our results are the first to demonstrate an edge-mediated relaxation of dietary restrictions in leaf-cutting ants. These findings render robust support to previous evidence indicating the reduction of bottom-up forces as a key factor explaining both edge-induced hyper-abundance and increased herbivory of leaf-cutting ants in human-modified Neotropical landscapes.

Leaf-cutting ants (species of Atta and Acromyrmex) are dominant herbivores and play a key role as ecosystem engineers of tropical and subtropical America (Fowler et al. 1989, Weber 1972). Not only are they among the most polyphagous and voracious herbivorous insects, cutting up to 15% y−1 of the leaf standing crop (Urbas et al. 2007, Wirth et al. 2003), but also they strongly affect the light environment and the nature of plant assemblages via ant-nest-mediated disturbances (Farji-Brener & Illes 2000, Hull-Sanders & Howard 2003, Moutinho et al. 2003). Some leaf-cutting ant species have turned into an omnipresent feature of present-day neotropical landscapes and a wealth of studies has documented their abundance to drastically increase with increasing agricultural land use, disturbance and deforestation/fragmentation (Fowler et al. 1986, Jaffe 1986, Terborgh et al. 2001, Vasconcelos & Cherrett 1995, Wirth et al. 2007). In view of their ecosystem engineering capacity and the ever-increasing conversion of tropical forests into agricultural landscapes (Wright 2005), it has been concluded that disturbance-driven accumulation of Atta colonies leads to far-reaching and deleterious consequences in present-day neotropical landscapes (Wirth et al. 2008). But what about the opposite scenario of regenerating forests? Is disturbance-mediated hyper-abundance of leaf-cutting ants a reversible phenomenon? We believe that this question is highly relevant because (1) knowledge of the dynamics of leaf-cutting ant populations during forest regeneration is lacking and (2) natural secondary succession has become a widespread phenomenon after land is abandoned or temporarily fallowed (Wright 2005). In the Brazilian Amazon during the 1990s, for example, secondary forests have reclaimed 31% of the once deforested land (Perz & Skole 2003).

Leaf-cutting ants (genera Atta and Acromyrmex) have been denoted key species of American rain-forest ecosystems (Fowler et al. 1989) because of their multifarious effects on the vegetation. Being dominant herbivores, cutting up to 13% of the standing leaf crop in a colony's territory per year, they affect directly and significantly individual plants, plant communities and ecosystems (Wirth et al. 2003). The considerable ecological impact of these ants is paralleled by the well-known fact that some species strongly benefit from human-driven habitat alterations and represent prime pests throughout Latin America (Cherrett 1986). Numerous studies have documented populations of leaf-cutting ant to increase with increasing agricultural land use, deforestation and/or disturbance (Fowler et al. 1986, Jaffe & Vilela 1989, Jonkman 1979). Specifically, elevated colony densities have been recorded in (1) transformed vegetation such as pastures (Fowler 1983) and plantations (Jaffe 1986, Oliveira et al. 1998), (2) early successional forests (Vasconcelos & Cherrett 1995), and recently (3) isolated forest remnants (Terborgh et al. 2001).