Sequential Sampling Plan of Anthonomus grandis (Coleoptera: Curculionidae) in Cotton Plants

Abstract

The boll weevil, Anthonomus grandis grandis Boheman (Coleoptera: Curculionidae), is one of the most important pests of cotton production worldwide. The objective of this work was to develop a sequential sampling plan for the boll weevil. The studies were conducted in Maracaju, MS, Brazil, in two seasons with cotton cultivar FM 993.A 10,000-m(2) area of cotton was subdivided into 100 of 10- by 10-m plots, and five plants per plot were evaluated weekly, recording the number of squares with feeding + oviposition punctures of A. grandis in each plant. A sequential sampling plan by the maximum likelihood ratio test was developed, using a 10% threshold level of squares attacked. A 5% security level was adopted for the elaboration of the sequential sampling plan. The type I and type II error used was 0.05, recommended for studies with insects. The adjustment of the frequency distributions used were divided into two phases, so that the model that best fit to the data was the negative binomial distribution up to 85 DAE (Phase I), and from there the best fit was Poisson distribution (Phase II). The equations that define the decision-making for Phase I are S-0 = -5.1743 + 0.5730N and S-1 - 5.1743 + 0.5730N, and for the Phase II are S-0 = -4.2479 + 0.5771N and S-1 = 4.2479 + 0.5771N. The sequential sampling plan developed indicated the maximum number of sample units expected for decision-making is similar to 39 and 31 samples for Phases I and II, respectively.