摘要
In this study, the multivariate tools, namely principal component analysis （PCA） and cluster analysis, were used to classify and measure the pattern of genetic diversity and evaluate the correlation of nine oil palm traits in 25 progenies. Fresh fruit bunch weight （FFB）, kernel to fruit （K/F） and kemel to bunch （K/B） ratios showed significant variance, while bunch number （BN）, kernel yield （KY） and oil yield （OY） showed little variance. Positive significant correlation between these traits and yield was appreciated through PCA, where 90.55% of the variation was explained by the first three principal components. Progeny grouping was performed and revealed three clusters of oil palm progenies. Cluster I contained progenies with high production of FFB, BN, OY and KY, while low height increment （HI） of palm trees was found in cluster II. However, most of progenies with high mean values of bunch spikelet weight （SpW）, average fruit weight （AFW）, K/F and K/B were grouped in cluster III. This grouping could help oil palm breeders to identify progenies with the traits of interest for breeding and commercial seed production.
