In the experimental design the plot is a determining factor for the success of the research. Therefore, it is necessary that the researcher adequately define the constitution of the plot or experimental unit, in order to increase the efficiency of the experiment and reduce experimental error. Due to the great variability in the plot size found in experiments with pineapple, a research was carried out to estimate the optimal size for plots considering different characteristics. The data used for this chapter were collected in 100 pineapple 'Vitória'. It was collected 100 fruits and 100 "D" leaves of plants grown in the same experimental area and subject to the same farming practices. It was evaluated in the fruits: length, diameter and weight. In the 'D' leaves it was measured the length and weight, and concentration of N, P and K. Estimates of plot size were based on the methods of linear response plateau and maximum curvature. The plot size for evaluation of pineapple 'Vitória' is 15-20 units per plot for the characteristics of the fruits and leaves by the method of linear model response plateau. To evaluate the concentrations of N, P and K, 10 leaves represent the optimum plot size by linear response plateau model. On the other hand, the plot size for evaluation of the fruit and leaves of the cultivar Vitória is from 1.7 to 2.8 units by the method of maximum curvature. While for the N, P and K, 3-5 units represent the experimental plot size by the method of maximum curvature.
Ananas comosus var. comosus; fruits; experimental unit
