The location and aggregation of gypsy moth egg masses in a population near New Paltz, N.Y., were monitored yearly during the period from 1971 to 1981. The population went from a peak in 1971, through a trough between 1974 and 1977, to another peak in 1981. Warm, dry conditions during May and June were associated with population increases, while cooler and wetter conditions were associated with decreases. During the years of greater abundance the egg mass density (no. per ha bark) varied with the abundance of xeric soils, xeric indicator tree species, and the preferred food plants. During the trough, however, the density of egg masses on vegetation was unrelated to woodland type. The ratio of egg masses found on human-built structures to those on vegetation varied inversely with the total egg mass count. During the trough and nontrough periods the density of egg masses on structures in a woodland depended on both the amount and types of structures present. Egg mass aggregation was determined at the following levels: microsite, individual plant, species, and landscape (woodland). In all cases aggregation was greatest during the trough. After the trough, landscape scale aggregation decreased sharply. By 1980 it attained a level similar to that of 1981 when the egg mass density was two orders of magnitude greater. The possible meanings of these changes in egg mass location and aggregation are explored in terms of gypsy moth survival strategies.