Abstract

The effects of forest fragmentation on beetle species composition were investigated in an experimentally fragmented tropical forest landscape in Central Amazonia. Leaf-litter beetles were sampled at seven distances from the forest edge (0–420 m) along forest edge-to-interior transects in two 100-ha forest fragments and two continuous forest edges, and at an identical series of distances along two deep continuous forest transects. Additional samples were taken at the centers of two 10-ha forest fragments and two 1-ha fragments. This sampling regime allowed discrimination between edge and fragment area effects. Beetle species composition changed significantly and independently with both decreasing distance from forest edge and decreasing fragment area. Edge effects on species composition were mediated by six important environmental variables: air temperature, canopy height, percent ground cover of twigs, litter biomass, litter moisture content, and an air temperature × distance from edge interaction effect, due to the different temperature profiles of edges with differing edge vegetation density. Population densities of 15 of the 32 most abundant beetle species tested (47%) were significantly affected by forest fragmentation. Species responses were classified empirically into four major categories: (A) edge sensitive, area insensitive; (B) area sensitive, edge insensitive; (C) edge and area sensitive; and (D) edge and area insensitive. Within these categories, trends in density were either positive (deep-forest species), or negative (disturbed-area species), with species showing the full spectrum of responses to fragmentation. The vast majority of species were adversely affected. Estimated species loss rates from forest fragments were: 49.8% of common species from 1-ha fragments, 29.8% from 10-ha fragments, and 13.8% from 100-ha fragments. Declining density was a significant precursor of species loss from forest fragments, but other species that did not show significant population density responses to fragmentation were also absent from some fragments, presumably by chance. The probability of species loss from forest fragments was not correlated with body size or trophic group for the 32 common species, although for the entire beetle assemblage (993 species) proportions of species in different trophic groups changed significantly with fragmentation. Rarity and population variability (in undisturbed forest) were significant predictors of susceptibility to fragmentation. Surprisingly, though, common species were significantly more likely to become locally extinct in small fragments than rarer species. This lends empirical support to models of multispecies coexistence under disturbance that suggest competitively dominant but poorly dispersing species are the first to become extinct due to habitat destruction. Thus, rarer species are predicted to be better dispersers and better at persisting.