Ecologists have long been intrigued by the factors that control the pattern of biodiversity, i.e., the distribution and abundance of species. Previous studies have demonstrated that coexisting species partition their resources and/or that the compositional similarity between communities is determined by environmental factors, lending support to the niche-assembly model. However, no attempt has been made to test whether the relative amount of resources that reflects relative niche space controls relative species abundance in communities. Here, we demonstrate that the relative abundance of butterfly species in island communities is significantly related to the relative biomasses of their host plants but not to the geographic distance between communities. In the studied communities, the biomass of particular host plant species positively affected the abundance of the butterfly species that used them, and consequently, influenced the relative abundance of the butterfly communities. This indicated that the niche space of butterflies (i.e., the amount of resources) strongly influences butterfly biodiversity patterns. We present this field evidence of the niche-apportionment model that propose that the relative amount of niche space explains the pattern of the relative abundance of the species in communities. neutral theory ͉ niche theory ͉ relative species abundance B iodiversity is often considered to be synonymous with species richness and relative species abundance (1). For more than half a century, ecologists have paid attention to the factors that determine the relative species abundance in ecological communities; i.e., the commonness and rarity of species (2). Host plants and their herbivore communities are good systems for examining how the diversity of herbivores is influenced by their resources; indeed, insect ecologists have focused considerable attention on the question of how host plant communities affect the species richness and composition of insects. Although a recent review (3) pointed out that little progress on this subject has been made since the 1980s, a number of important studies have demonstrated that a large number of plant species is frequently correlated with a large number of insect species (4-6). Experimental studies have also demonstrated a positive relationship between the diversity of plant species and the diversity of consumers (7-9). The abundance of resources also represents an important factor in the structuring of insect communities (10). A number of studies have demonstrated that resource abundance explains the variation in the abundance and species richness of herbivorous insects (11)(12)(13)(14). However, these studies on host plant and herbivore communities have treated species richness and species composition as components of the diversity, and no studies have examined the relative abundance patterns of both host plants and herbivores.On the other hand, recent theoretical arguments have focused on whether the neutral or niche theory better explains the patterns of relative spec...