The objective of this work was to estimate the minimum harvest number in experiments with cucumber. Data of an experiment comparing 14 japanese cucumber hybrids in the randomized block design, with four replications and five plants per plot, under protected cultivation, were analyzed. Thirty-two harvestings were made, three harvests a week, during 72 days to estimate repeatability coefficient based on analysis of the principal components for number and weight of total and commercial fruits per plant. Results allowed to conclude that less than nine harvestings were enough to verify yield differences among hybrids, with 95% of probability to identify the highest yielding ones.
Cucumis sativus L.; yield; harvesting week
