Few studies were conducted so far in regard to the use of row borders and number of replications to define an experimental plan. Therefore an experiment was conducted in order to estimate the number of replications necessary to substitute greater row numbers of the plot for the use on row borders. Two corn hybrids were used and eight treatment in which plots with two, three, four and five rows with and without two lateral row borders and two and three row plots with a single border row. There was neither difference among the estimated yield average of the hybrids and the different plots and nor interaction between the plot forms and hybrids as well as homogeneity of variance estimates among the plots of the two hybrids. In hybrid corn yield experiments there is no advantage to substitute the use of row borders for greater number of replications. However, a greater number of replications such as eight replications and two row plots with no border rows, will result in more precise experimental plans.
Zea mays; experimental design; precision
