Abstract
The purpose of sampling sugar-cane froghoppers in Trinidad is to determine whether or not an infestation requires insecticidal control. Two sequential schemes, based on the Poisson distribution, were computed for the northern and southern parts of the cane-belt where different spraying-thresholds apply. Decisions on whether or not to spray based on sequential sampling were shown to agree closely with those based on random sampling even when the number of cane-stems sampled sequentially was arbitrarily limited to half that sampled randomly. Sequential sampling, therefore, appeared to be a promising supplement or alternative to the currently used random sampling system.
