Abstract
Fruit maturing time (FMT), fruit weight (FW), and soluble solids content (SSC) in 15 identical persimmon (Diospyros kaki Thunb.) cultivars were evaluated at Akitsu, Hiroshima of Japan and Meixian, Shaanxi of China. The data were subjected to two-way analysis of variance, with the factors of location and cultivar. This model of analysis-of variance was applicable because the residue estimates approached a normal distribution as determined by the Kolmogorov-Smirnov one-sample test. FMT and FW were greatly and significantly affected by location and cultivar (P<0.01), whereas SSC was not for location (P>0.05). Coefficients of correlation between the two locations were 0.79**, 0.89**, and 0.52* for FMT, FW, and SSC, respectively. The cultivars evaluated at Akitsu, on the average, matured 18 days later and weighed 70g more than did those at Meixian. The mean values for SSC differed little between the two sites. These results indicated that cultivar performance in one location could be estimated from that in another location by incorporating the correction constant for the mean differences in FMT and FW between locations.
