The objectives of this research were to determine the optimum plot size and the number of repetitions to evaluate characters of jack bean (Canavalia ensiformis). It was carried six uniformity assays of size 5m×5m (25m2). Each assay was divided in 25 basic experimental units (UEB) of 1m×1m, totaling 150UEB. The fresh mass of pods (MVV) and fresh mass of aerial part without pods (MVPASV) were weighed and fresh mass of the aerial part (MVPA=MVV+MVPASV) of plants of each UEB was calculated. The optimum plot size was determined and the means were compared, between characters MVV, MVPASV and MVPA. The number of replicates was determined. The optimum plot sizes to evaluate the fresh mass of pods, fresh mass of aerial part without pods and fresh mass of aerial part are, respectively, 8.59m2, 6.14m2 and 5.85m2. Four replications, to evaluate up to 50 treatments, in randomized block designs, are sufficient to identify, as significant at the 5% probability by Tukey test, differences between treatment means 55.48%, 39.66% and 37.78% of the average fresh mass of pods, fresh mass of aerial part without pods and fresh mass of aerial part of the experiment, respectively
Canavalia ensiformis; experimental design; experimental precision
