The control of the flat mite Brevipalpus phoenicis (Geijskes, 1939), vector of the citrus leprous virus, must be done when its population reach the threshold level, estimated by the monitoring the mite population density. The objective of this study was to determine the acceptable sample size for estimating the flat mite population, for later decision. The experiment was realized at the Cambuhy Farm, Matão - SP. It was casually select an orchard of Valência variety, eight years of age, planted in a spacing of 7 x 3.5m, with 2.480 plants. This orchard was inspected 1, 2, 3, 5, 10 and 100% of plants, respectively 25, 50, 74, 124, 248 and 2480 plants, by walking in the direction of the planting lines. It was inspected 3 fruits or branch per plant. According to the results, it was observed that the error percentage for estimating the mean percentage of fruits with flat mite, in actual situation of inspection of flat mite leprous in citrus, when it was inspected only 1% of plants (25 plants) the error is 50%, however, for an infestation of 10% the variation of infested fruits percentage was between 5 and 15%, leading the farmers to under or super estimate the mite population density, increasing the costs with unnecessary pulverization or an inefficient control of the flat mite. To the error in the sample remains in an acceptable situation (20 to 30% of error), 105 plants must be inspected. In fruits percentage with more than 10 flat mites, it was verified that for the acceptable situation (20 to 30%), 540 plants must be inspected.
sampling; flat mite; citrus
