Planejamento experimental na cultura da berinjela

Abstract

Eggplant consumption has increased in recent years, boosted by its multiple benefits to human health. Research should be carried out with this horticultural species in order to provide technical recommendations for increasing fruit production and quality. In this sense, the objectives of this work were to determine the influence of the size of the uniformity trial on the plot size estimation and to estimate the sample size, plot size and number of replications sufficient for trials with eggplant culture. Two uniformity trials were performed with eggplant in plastic tunnel. The fresh fruit mass and the number of fruits were evaluated in each basic experimental unit and in each harvest. Crops were analyzed individually and pooled. In each tunnel and variable, 25 trials of uniformity of different sizes were planned per crop and crop row (individual and grouped). For each planned uniformity trial, 3,000 bootstrap resampling was performed with replacement. In each planned uniformity trial, 3,000 plot sizes were estimated by the maximum curvature of the coefficient of variation method and the mean and 95% confidence interval were calculated by the difference between the 97.5% and 2.5%. The heterogeneity was found between the harvest rows and the individual crops in the same row. However, when harvests were pooled, this fact was not observed. The size of the uniformity trial influences the plot size estimation because trials with low number of basic experimental unit’s present high experimental variability and inaccurate estimates. When the harvests are grouped, there was a reduction in plot size and the required sample, the clustering of six harvests presented the smallest sample size and plot size. In order to estimate the plot size in experiments with eggplant, uniformity trials are required composed of 16 basic experimental units to evaluate the fresh fruit mass variable and 14 basic experimental units for the variable number of fruits. The plot size is five plants, the sample is nine plants in the direction of the harvest row with a confidence interval of 20%. In order to reach a significant minimum difference in the Tukey test of 30% of the treatment, three replications with five plants per plot are required.