This research was aimed at evaluating the relationship between one plot coefficient variation and heterogeneity index with the optimal potato plot size. Tuber yield data was gotten from 12 potato homogeneity trials. Each trial had 12 rows and 24 hills, from which 18 plot sizes, differing in row and hill numbers, were designed. The functions CV(x)=A/X B and VU(x)=V1/Xb were adjusted, where X is the plot size in number of basic units (BU), CV(x) is the coefficient of variation between plots with X BU, and VU(x) is the BU variation between plots with X BU. Optimal plot size estimated by three different methods and the results were correlated with the parameter estimations of the above two functions. Based upon path analysis, the variation coefficient between plots of the basic unit affects more the estimation of optimal plot size than the heterogeneity index in potato uniformity trials.
Solanum tuberosum L.; heterogeneity index; experimental design; experimental precision
