Good experimental planning is essential in order to obtain reliable and good results. In this context, the correct choice of plot size and form, which depends on the kind of crop and environment, deserves special attention. This research was aimed at estimating the optimal plot size as a function of the experimental precision, in a sorghum yield (variety ‘BR 304’) experiment in two sowing seasons. Data of the experiment were studied during the 2005/2006 growing season at the Universidade Federal de Santa Maria. Three areas were sown in two dates, (12/7/2005 and 01/05/2006) in a completely randomized design with three replications. Each area was formed by four rows 0.5 m apart, totaling 36 plants per row. These 144 plants were considered as basic units in order to simulate different plot sizes. It was concluded that for sorghum yield experiments, the optimal plot size is formed by eight plants or approximately or one linear meter of row and greater precision can be obtained by increasing the number of replications regardless the number of treatments and sowing dates.
Sorghum bicolor L.; basic unit; experiment planning
