Potato trial dimensions and optimum plot size estimations

The objective of this work was to assess the relationship among experimental area in uniformity trial dimensions, optimum plot size estimations and experimental precision of potato trials. A homogeneity trial was carried out at the production site of Fundação Estadual de Pesquisa Agropecuária, Júlio de Castilhos, RS, Brazil. Tuber yield of 3,456 hills (24 rows of 144 hills) was quantified and data were used for planning ten different size trials, in which the number of rows and hills per row varied. Estimations of optimum plot size and differences between means of treatments were done for each trial. Based on linear correlation and path analysis, the size of potato trial using different hill numbers does not affect the estimation of soil heterogeneity index and the estimation of optimum plot size. Experimental areas with high number of hills per row can be used to improve precision of potato yield evaluations. One row with six hill plots and treatments distributed in the row, taken as block, increase experimental precision in potato.