*Central Statistical Organisation, New Delhi.
Institute of Agricultural Research Statistics, New Delhi-110012
The statistical analysis of data on apples collected from Summer Ford Orchard (Nainital) during 1965 and 1969 have shown that the coefficient of variation decreases with the increase in the plot size both with and without blocking arrangements of plots. Single tree plots were found to be most efficient. Fairfield Smith law was found to be satisfactory for describing the relationship between plot size and the variance of the means per tree.
It was observed that for 20 per cent standard error, the number of replications decreases with the increase in the size of the plot for all block sizes, however, the total experimental material required was minimum when a single tree experimental unit was adopted. With the increase in the block size, efficiency of the experiment decreases for all plot sizes as generally expected.