Part I. The microspatial distribution of pupal green cloverworms, Plathypena scabra (F.), was studied intensively in 6 soybean fields during 1979 and 1980. Pupae were recovered from the plant canopy (10%) and soil litter surface (90%) but never from subsurface soil layers. Pupal densities were biased with respect to directional aspect and distance from the soybean row. Litter cover and population density influenced pupal distribution;Part II. Intensive surveys of GCW pupae were conducted in 6 soybean fields during 1979 and 1980 to develop a sampling program suitable for life table research. Nested analyses of variance (NANOVA) demonstrated that differences within soil-litter sample sites and among plant-canopy samples accounted for > 95% of the total variation observed in pupal densities on the soil surface and in the canopy, respectively. Simple random sampling plans are satisfactory for pupal surveys. A 50 x 60-cm area, centered over the soybean row, represented the best compromise between sampling cost and precision. Sequential count plans, based on Taylor's Power Law, were calculated for estimating pupal densities;Part III. Six partial life tables were prepared for pupal green cloverworms, Plathypena scabra (F.), during 1979 and 1980. Pupal dynamics were characterized by large 1st generation densities that declined below detectable levels during generation 2 in 1979, and by small 1st generation densities that expanded 7-fold during generation 2 in 1980. The 1979 and 1980 density patterns are characteristic of outbreak and endemic GCW population configurations, respectively. Pupal survivorship was 4-fold greater under endemic population levels than under outbreak levels during generation 1. Pupae were parasitized by 11 primary species. Total parasitization represented less than 18% of total 1979 pupal mortality but accounted for 60% of total 1980 mortality. Infections caused by a microsporidan and the entomogenous fungus, Nomuraea rileyi (Farlow) Sampson, accounted for a small portion of total mortality, as did deaths attributed to nonviability. Predation was an important source of pupal mortality and was 2.5 times greater under outbreak levels than under endemic levels during generation 1 (i.e., 64% mortality vs 24.1% mortality). A modified key factor analysis indicated that no single mortality component was correlated with changes in total pupal mortality. Regression analyses generally failed to detect density-dependent mortality factors, but did indicate that total pupal mortality was density-dependent.