ABSTRACT
Optimum plot size is of fundamental importance for the results, precision and cost-effectiveness of experiments. The aim of this study therefore, was to estimate optimum plot size in a strain of the castor bean by means of the Linear Response Plateau Model (LRPM). The experiment consisted of a test for uniformity, carried out in Cruz das Almas, in the State of Bahia, Brazil. The UFRB-222 strain was used, which was planted at a spacing of 3 m x 1 m, in 10 rows of 37 plants each. The eight central rows were considered, each with 35 plants per row giving a total of 280 plants and a working area of 840 m2. For the evaluations, each plant was taken as one basic unit (bu) with an area of 3 m2, resulting in 280 bu. The characteristics under evaluation were the number of racemes harvested, the mean length of the first three racemes, number of seeds per raceme, seed weight per raceme, raceme weight per plant, number of seeds per plant and productivity. For productivity and its components, the optimum plot size is 11 experimental units in an area of 33 m2, at a spacing of 3 m between rows and 1 m between plants.
Key words:
Ricinus communis L.; Plateau model; Experimental accuracy; Uniformity
