Varietal Differences in Fruit Bearing in Japanese Persimmon (Diospyros kaki Thunb.) and their Yearly Fluctuations

Fruit bearing and its yearly fluctuation were examined in 17 varieties of Japanese persimmon (Diospyros kaki Thunb.) to develop a new, effective method for selecting progenies with good bearing habits in the breeding program. About 50 flowers, arranged to a leaf-flower ratio of 15:1, were hand pollinated or covered with bags to protect against cross pollination, and the number of fruit set from the flowers was counted at harvest. The experiment was repeated with 12 varieties for 3 seasons (1981-1983).Fruit drop was confined to the early developmental stages in all varieties except for‘Okitsu No. 15’which showed a heavy late drop. Generally, yearly fluctuation of fruit set was small in pollinated fruits. On the contrary, there were large fluctuations in non-pollinated fruits; for example, from 9% to 84% in‘Tenjingosho’. Parthenocarpic fruit set of non-pollinated fruits, however, changed yearly in parallel in every variety. This seemed to be due to the genotype of each variety.Yearly fluctuation in seed number of pollinated fruits at harvest time was very small; thus there was a high correlation between yearly seed number and the mean of 3 seasons. Although there was no correlation between fruit set and seed number on the whole varieties, varieties with low parthenocarpic ability showed a high correlation between them in 1981, when fruit set was comparatively low in all varieties. In many cases, seed number seemed to be a useful index to represent the ability to set seed at an early stage. This ability can be accurately estimated within a few years by the use of this index, because its yearly variation is very small.Fruit-bearing characteristics of seedling progenies in the breeding program could be predicted by the above test of pollinated and non-pollinated fruit, as compared with standard varieties.