An international project, DIWPA-IBOY, took place for simultaneously observing biodiversity throughout the Western-Pacific and Asian regions in 2001-2003, as one of the core projects for International Biodiversity Observation Year, a crosscutting network activity of DIVERSITAS (an international programme of biodiversity science). DIWPA-IBOY provides extensive data on species diversity obtained by the standardized method. Under this project, 51,742 individuals of Lepidoptera and 11,633 of Coleoptera were collected by light traps from the Tomakomai Experimental Forest of Hokkaido University, one of the core DIWPA-IBOY sites, in the cool-temperate region of northern Japan. Based on these data, this study examined the relative abundance distribution (RAD) to evaluate the amount of rare species in the Lepidoptera and Coleoptera communities. The beta diversities between sampling seasons, forest strata, and trap sites were also assessed to evaluate the spatio-temporal variability of species composition in these communities. In the analysis of the RAD, the best-fit model was selected from the log-Normal, Zipf-Mandelbrot, and Zipf models differing in the tail length of the RAD, i.e., the proportion of rare species. To explore the beta diversity between samples, the abundance-based Jaccard index with an unseen species estimator was calculated, and then a hierarchical clustering analysis was conducted. As a result of RAD analysis, the Coleoptera community was regarded as containing a larger proportion of rare species than the Lepidoptera community. The seasonal compartmentalization of the community, deduced from the beta-diversity analysis, was finer in Lepidoptera (seven assemblages recognized) than in Coleoptera (three assemblages). The spatial (vertical and horizontal) compartmentalization was negligible in both communities. The coincidence of the larger proportion of rare species and the lower beta diversity between seasons in the Coleoptera community was explained by the longer life spans of beetles compared to moths, based on the assumption that the length of life span acts as a temporal agent for mass effect on the analogy of the migration rate as a spatial agent for mass effect.