Abstract
A sequential sampling system for early-instar Douglas-fir tussock moth larvae based on beating 3 lower branches from each of a number of trees is described. The relationship between lower- and mid-crown insect densities is given. The number of samples required to assess moderate insect densities using a fixed level of precision and at low insect densities using a critical density level is presented. The system is designed to be used during the early stages of an outbreak in stands not previously defoliated.