Sequential sampling plan to evaluate the infestation of the coffee leaf miner (Lepidoptera:Lyonetiidae) in southern Minas Gerais state, Brazil

Abstract

This work had the aim of constructing a sequential sampling plant to the coffee leaf miner, Perileucoptera coffeella (Guérin Mènéville, 1842). Two coffee fields, located in southern Minas Gerais State, Lavras and São Sebastião do Paraíso, were sampled every two weeks for a year to evaluate coffee leaf miner infestations. Each sampling consisted of ten plants, taken randomly at high sites of each area. Sixty leaves were collected from each plant: 20 at the lower one-third, 20 at the medium one-third, and 20 at the upper one-third. The mean and the variance were fitted to the linear regression of Lloyd's mean crowding m* on mean m, and to Taylor's power law. The former model fitted better the data, and therefore it was used to construct the sequential sampling plan, by the confidence interval of Iwao (1975). A sampling table is presented in order to simplify sampling procedure.