THE SICILIAN OLIVE GERMPLASM AND ITS CHARACTERIZATION BY USING STATISTICAL METHODS
Fruit, pit and leaf biometric characters were evaluated for their ability to discriminate olive cultivars. Recordings from 25 indigenous Sicilian cultivars were used. Data showed a large amount of variability among the accessions that can be considered as a valuable source of genetic traits for breeding programmes. Discriminant functions calculated using fruit characters classified correctly 87% of the observations but only less than 50% using leaf and leaf colour. No identities between cultivars were detected. Principal components analysis on the between-cultivars correlation matrix for all variables extracted 6 unrelated variables explaining >95% of total variance and showed few patterns of similarities between cultivars.
Barone, E., Di Marco, L., Motisi, A. and Caruso, T. (1994). THE SICILIAN OLIVE GERMPLASM AND ITS CHARACTERIZATION BY USING STATISTICAL METHODS. Acta Hortic. 356, 66-69