To determine the effect of rivers, environmental conditions, and isolation by distance on the distribution of species in Amazonia. Location: Brazilian Amazonia. Time period: Current. Major taxa studied: Birds, fishes, bats, ants, termites, butterflies, ferns + lycophytes, gingers and palms. We compiled a unique dataset of biotic and abiotic information from 822 plots spread over the Brazilian Amazon. We evaluated the effects of environment, geographic distance and dispersal barriers (rivers) on assemblage composition of animal and plant taxa using multivariate techniques and distance- and raw-data-based regression approaches. Environmental variables (soil/water), geographic distance, and rivers were associated with the distribution of most taxa. The wide and relatively old Amazon River tended to determine differences in community composition for most biological groups. Despite this association, environment and geographic distance were generally more important than rivers in explaining the changes in species composition. The results from multi-taxa comparisons suggest that variation in community composition in Amazonia reflects both dispersal limitation (isolation by distance or by large rivers) and the adaptation of species to local environmental conditions. Larger and older river barriers influenced the distribution of species. However, in general this effect is weaker than the effects of environmental gradients or geographical distance at broad scales in Amazonia, but the relative importance of each of these processes varies among biological groups.

The spatial distribution of species is affected by dispersal barriers, local environmental conditions and climate. However, the effect of species dispersal and their adaptation to the environment across geographic scales is poorly understood. To investigate the distribution of species from local to broad geographic scales, we sampled termites in 198 transects distributed in 13 sampling grids in the Brazilian Amazonian forest. The sampling grids encompassed an area of 271 500 km2 and included the five major biogeographic regions delimited by Amazonian rivers. Environmental data for each transect were obtained from local measurements and remote sensing. Similar to previous studies, termite species composition at the local scale was mostly associated with measures of soil texture and chemistry. In contrast, termite species composition at broad geographic scales was associated with soil nutrients, and the geographic position of the transects. Between 17 and 30% of the variance in termite species composition could be attributed exclusively to the geographic position of the transects, but could not be attributed to measured environmental variables or the presence of major rivers. Isolation by distance may have strong effects on termite species composition due to historic processes and the spatially structured environments along distinct geological formations of Amazonia. However, in contrast to many taxa in Amazonia, there is no evidence that major rivers are important barriers to termite dispersal.

No abstract

This study investigated the spatial distribution of an Amazonian fruit-feeding butterfly assemblage by linking species taxonomic and functional approaches. We hypothesized that: 1) vegetation richness (i.e., resources) and abundance of insectivorous birds (i.e., predators) should drive changes in butterfly taxonomic composition, 2) larval diet breadth should decrease with increase of plant species richness, 3) small-sized adults should be favored by higher abundance of birds, and 4) communities with eyespot markings should be able to exploit areas with higher predation pressure. Fruit-feeding butterflies were sampled with bait traps and insect nets across 25 km(2) of an Amazonian ombrophilous forest in Brazil. We measured larval diet breadth, adult body size, and wing marking of all butterflies. Our results showed that plant species richness explained most of the variation in butterfly taxonomic turnover. Also, community average diet breadth decreased with increase of plant species richness, which supports our expectations. In contrast, community average body size increased with the abundance of birds, refuting our hypothesis. We detected no influence of environmental gradients on the occurrence of species with eyespot markings. The association between butterfly taxonomic and functional composition points to a mediator role of the functional traits in the environmental filtering of butterflies. The incorporation of the functional approach into the analyses allowed for the detection of relationships that were not observed using a strictly taxonomic perspective and provided an extra insight into comprehending the potential adaptive strategies of butterflies.

Resources influence population growth, interspecific interactions, territoriality and, in combination with moisture content, affect terrestrial arthropod distribution and abundance. Ants are usually described as interactive and compete in transitive hierarchies, where the dominants behaviourally exclude subordinate species from food resources. In this study, we evaluated the effects of (i) dominant ants, soil moisture and an artificial resource gradient on the number of ant species attracted to baits; and (ii) how soil moisture and an artificial resource gradient change the number of controlled baits in a Central Amazonian rain forest. We sampled 30 100-m-long transects, located at least 200 m apart. The transects were established with six different bait densities varying between six and 41 baits and the soil moisture content was measured at 10 points for each transect. Six ant species were considered dominant, and had negative correlations with the number of species at baits (r 2 = 0.186; F1,28 = 6.419; P = 0.017). However, almost half of the transects showed low abundance of dominant species (<30%), and relatively high number of species (mean of 20.1 Ϯ 8.75). Resource availability and soil moisture had negative and positive correlations, with number of controlled baits. These results suggest that, even though the dominance is relatively poorly developed on the floor of this tropical forest, both resource availability and soil moisture affect resource control, and thus, the number of species that use baits.

2 Programa de P os-Graduac ßão em Entomologia, INPA, CP 2223, CEP 69080-971, Manaus, AM, Brazil 3 Programa de P os-Graduac ßão em Ecologia, INPA, CP 2223, CEP 69080-971, ABSTRACTArthropod abundance and diversity are remarkable in tropical forests, but are also spatially patchy. This has been attributed either to resources, predators, abiotic conditions or disturbances, but whether such factors may simultaneously shape arthropod assemblage structure is little known. We used cockroaches to test for multiple environmental controls on assemblage structure in 25 km 2 of Amazonian forest. We performed nocturnal, direct searches for cockroaches in 30 plots (250 m 9 2 m) during two seasons, and gathered data on biotic and abiotic factors from previous studies. Cockroach abundance increased with dry litter mass, a measure of resource amount, while species richness increased with litter phosphorus content, a measure of resource availability. Cockroach abundance and species richness decreased with ant relative abundance. Cockroach species composition changed along the gradient of: (1) soil clay content, which correlates with a broad differentiation between flood-prone and non-flooded forest; (2) soil relative moisture, consistent with known interspecific variation in desiccation tolerance; and (3) according to the abundance of ants, a potential predator. Turnover in species composition was correlated with abiotic conditions-sorting species according to physiological requirements and to disturbancerelated life history traits-and to ants' selective pressure. Cockroach abundance, diversity, and composition seem to be controlled by distinct sets of environmental factors, but predators which were represented by ants, emerged as a common factor underlying cockroach distribution. Such patterns of community structure may have been previously overlooked by undue focus on single or a few factors, and may be common to tropical forest arthropods.Abstract in Portuguese is available in the online version of this article.

Predation is a key determinant of prey community structure, but few studies have measured the effect of multiple predators on a highly diverse prey community. In this study, we asked whether the abundance, species richness, and species composition of a species-rich assemblage of termites in an Amazonian rain forest is more strongly associated with the density of predatory ants or with measures of vegetation, and soil texture and chemistry. We sampled termite assemblages with standardized hand-collecting in 30 transects arranged in a 5 km 9 6 km grid in a terra firme Amazonian rain forest. For each transect, we also measured vegetation structure, soil texture, and soil phosphorus, and estimated the density of predatory ants from baits, pitfall traps, and Winkler samples. Seventy-nine termite species were recorded, and the total density of predatory ants was the strongest single predictor of local termite abundance (r = À0.66) and termite species richness (r = À0.44). In contrast, termite abundance and species richness were not strongly correlated with edaphic conditions (|r| < 0.01), or with the density of non-predatory ants (r abund = À0.27; r s = À0.06). Termite species composition was correlated with soil phosphorus content (r = 0.79), clay content (r = À0.75), and tree density (r = À0.42). Assemblage patterns were consistent with the hypothesis that ants collectively behaved as generalist predators, reducing total termite abundance, and species richness. There was no evidence that ants behaved as keystone predators, or that any single termite species benefited from the reduction in the abundance of potential competitors.

1. Understanding the causes of the spatial variation of biodiversity is an important goal in community ecology. This study investigated the response of fruit‐feeding butterfly assemblages to environmental gradients resulting from the transition from ombrophilous forests to white‐sand forests in northern Brazil by assessing taxonomic, functional and phylogenetic descriptors of community structure. 2. Butterflies were sampled with bait traps in the Brazilian Amazon, while their traits and phylogenetic relatedness were either measured directly or gathered from the literature. Then, the effects of plant species turnover, light intensity, and flood risk on butterfly community patterns were tested. 3. Butterfly abundance increased with light intensity in non‐forested white‐sand habitats, whereas ombrophilous forests harboured higher species richness. A trade‐off was observed between richness and abundance across the gradient. Plant turnover strongly drove shifts in butterfly species composition. In white‐sand habitats, butterflies had higher dispersal capacity, faster flights and generalist larvae, whereas in periodically flooded plots the functional diversity of flight‐related traits was higher. No phylogenetic response to environmental gradients was detected. 4. These results exposed the complexity of community structure across a mosaic landscape and how analysing more than one dimension of biodiversity may reveal underlying relationships not apparent under isolated approaches. Caution is required when using phylogeny as a proxy for functional similarity, as the responses were not congruent in this study. The ecological patterns revealed here, aligned with further regional‐scale studies, can provide complementary perspectives on butterfly spatial distribution and ensure that appropriate conservation policies are developed.