Estimating apple fruit size distribution from early-season fruit diameter measurements

In three experiments, diameters of apples representing 7% to 30% of the fruit on a tree were measured at ≈60 days after full bloom. Using previously published regression equations, the early-season fruit diameter values were used to estimate apple fruit weight at harvest (FWH). At harvest, all fruit on sample trees were weighed and the distributions of estimated FWH for fruit measured early in the season were compared with distributions of the actual FWH for whole trees. Actual FWH was normally distributed for only one of the three experiments. Although the estimated mean FWH averaged for the 10 trees was within 9% of the actual mean FWH for all three experiments, the distribution of estimated FWH differed significantly from the actual distribution for all three experiments. All fruit were then assigned to appropriate commercial fruit sizes or box counts (number of fruit/19.05 kg). Fruit size tended to peak on the same four box counts for the estimated and actual populations, but the estimated populations had too few fruits in the small- and large-size box counts. Using early-season estimates of FWH, commercial apple growers and packers can predict fairly accurately the percentage of the crop that will fall into the peak box counts, but a more accurate early-season estimate of the fruit size distribution will likely require measuring 50% of the fruit on a tree.