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).

The use of indicator taxa as biodiversity surrogates has received widespread attention in conservation planning, but remains a highly contentious issue. Here we assess biodiversity surrogacy in the two most important biomes of tropical northeastern Brazil, Atlantic forest and Caatinga. We examine the extent to which species richness is correlated among taxonomic groups, and assess relationships between the richness of individual groups, and combinations of groups and total species richness. We introduce a new approach to dealing with autocorrelation between focal taxon richness and total species richness, using standardized data such that each taxon is given equal weight. Our Atlantic forest data covered seven taxa (bryophytes, pteridophytes, trees, ants, euglosine bees, birds, and mammals; total of 768 species) sampled from 12 sites; in Caatinga it was four taxa (trees, spiders, beetles and ants; total of 184 species) from 25 sites. Our results showed that:(1) in nearly all cases the species richnesses of individual taxa were significantly correlated with each other; (2) the species richnesses of most individual taxa were significantly correlated with total species richness in both biomes; (3) only two taxa were required for excellent (R 2 [ 80%) surrogacy of total species richness in both biomes; and (4) the same two taxa (trees and ants) can provide reasonable (R 2 [ 60%) surrogacy for total richness in these contrasting biomes. Our findings therefore suggest that the 'shopping basket of taxa' required for effective biodiversity surrogacy may not only need to be very small (two taxa), but may also be very limited in composition.

Major shifts in the availability of palatable plant resources are of key relevance to the ecology of leaf‐cutting ants in human‐modified landscapes. However, our knowledge is still limited regarding the ability of these ants to adjust their foraging strategy to dynamic environments. Here, we examine a set of forest stand attributes acting as modulating forces for the spatiotemporal architecture of foraging trail networks developed by Atta cephalotes L. (Hymenoptera: Formicidae: Attini). During a 12‐month period, we mapped the foraging systems of 12 colonies located in Atlantic forest patches with differing size, regeneration age, and abundance of pioneer plants, and examined the variation in five trail system attributes (number of trails, branching points, leaf sources, linear foraging distance, and trail complexity) in response to these patch‐related variables. Both the month‐to‐month differences (depicted in annual trail maps) and the steadily accumulating number of trails, trail‐branching points, leaf sources, and linear foraging distance illustrated the dynamic nature of spatial foraging and trail complexity. Most measures of trail architecture correlated positively with the number of pioneer trees across the secondary forest patches, but no effects from patch age and size were observed (except for number of leaf sources). Trail system complexity (measured as fractal dimension; Df index) varied from 1.114 to 1.277 along the 12 months through which ant foraging was monitored, with a marginal trend to increase with the abundance of pioneer stems. Our results suggest that some leaf‐cutting ant species are able to generate highly flexible trail networks (via fine‐tuned adjustment of foraging patterns), allowing them to profit from the continuous emergence/recruitment of palatable resources.