Modelos de regressão com platô na estimativa do tamanho de parcelas em experimento de conservação in vitro de maracujazeiro

Abstract

The determination of the plot size is a practical question to the experimental design, and its characterization in an optimized way allows obtaining larger precision and quality results. This research aimed to determine the plot size in experiments in vitro that seek the passion fruit plant conservation in ten uniformity assays with the species Passiflora Giberti N. E. Brown. The tests of uniformity came from an experiment conducted in a completely randomized design with treatments in a factorial design with three concentrations of sucrose, three concentrations of sorbitol and a control. Each treatment was considered as a uniformity assay, with 20 basic units. The evaluations of the experiments were done in the 60 days of incubation, being measured the shoots' length. For the estimate of the optimal plot size it was used the models segmented with linear- plateau, and the quadratic-plateau. The plot size varied with the method used, with values from 6 basic units for the linear-plateau model segmented, and from 10 basic units, for quadratic-plateau model.