Infodoc : Sciences, Techniques et Culture

Samples representing 32 624 arthropods were collected using standard sampling methods from five habitats in lowland rain forest (canopy, tree trunks, low herb layer, leaf litter and soil) in Seram, Indonesia. The relationships between arthropod abundance (number of individuals), body size, and biomass are described for these five habitats, for the predicted arthropod fauna of a hectare of forest, and for different trophic groups. The results are similar for all five habitats: abundance decreases and biomass increases with body size. Broadly similar principles apply for herbivores, decomposers and predators but not for "tourists". Thus the "equal biomass hypothesis", which predicts roughly constant biomass across biomass logarithmic size classes, is rejected. Previous studies have sugg[...]

Samples representing 32 624 arthropods were collected using standard sampling methods from five habitats in lowland rain forest (canopy, tree trunks, low herb layer, leaf litter and soil) in Seram, Indonesia. The relationships between arthropod abundance (number of individuals), body size, and biomass are described for these five habitats, for the predicted arthropod fauna of a hectare of forest, and for different trophic groups. The results are similar for all five habitats: abundance decreases and biomass increases with body size. Broadly similar principles apply for herbivores, decomposers and predators but not for "tourists". Thus the "equal biomass hypothesis", which predicts roughly constant biomass across biomass logarithmic size classes, is rejected. Previous studies have suggested that the fractal nature of surfaces is important in determining abundance-body size relationships in animal communities. These predict slopes of approximately - 3.25 for log abundance against log body size. In this study the steepest slope was only - 2.43. This suggests that the fractal nature of surfaces are less important in determining arthropod assemblages than has been considered. Two factors may account for the difference in predicted and observed slopes. First, the assumed metabolic rate component in fractal arguments (i.e. that metabolic rate scales with body size to the 0.75 power) is wong and second smaller organisms are under-sampled from all habitats. Under-sampling of smaller organisms may also account, at least in part, for the observed slopes which result in our rejection of the "equal biomass hypothesis". "(Résumé des auteurs)".