Phenotypic Selection to Avoid Discarding Target Genotypes for Four Fruit Traits Based on Environmental Variances in a Pineapple Cross-seedling Population

Early ripening [earlier than 1 Aug during fruit harvest time (FHT)], large fruit weight (FW; >1000 g), high sugar content [>17% soluble solids content (SSC)], and low acidity in fruit juice (<0.7%) are important breeding targets of pineapple for table fresh fruit use in Japan. We investigated the efficiency of primary selection based on the four fruit traits using 129 first-fruiting F1 offspring population of ‘Yugafu’ × ‘Yonekura’ without replicates. Separately, environmental variances were estimated by an analysis of variance using evaluation data from 50 or 49 offspring in three replicates and two-year repeats. The phenotypic distribution in the 129 F1 population approached a normal distribution (P > 0.05). The genotypic distribution was obtained as a normal distribution with the population mean as the mean and genotypic variance obtained by subtracting the environmental variance from the phenotypic variance. The target genotypes were estimated at 14.4%, 58.7%, 5.0%, and 50.0% of the F1 population for FHT, FW, SSC, and acidity, respectively. Critical phenotypic values were established as the upper (FHT and acidity) and lower (FW and SSC) limits of the critical genotypic values at the 95% probability level. The phenotypic selection was made based on the critical phenotypic value, resulting in 45.0%, 88.4%, 27.1%, and 79.1% of the offspring selected for FHT, FW, SSC, and acidity, respectively, and 12.4% simultaneously for all four fruit traits. The results showed that the phenotypic primary selection reduced the population size to 12.4%, avoiding the discarding of target genotypes with a low risk. If breeders intend to further reduce the population size, then increasing the number of traits subject to primary selection would be effective.