Study on Traits Related to Seed Yield in Safflower by Factor Analysis

Study on genetic variation is an important feature in plant breeding. Knowledge on genetic variation of plant populations enables the breeders to select better methods for breeding of their materials. The present study was carried out in a lattice design in Sararood Dryland Research Station during 2001 to investigate variations of different traits in 56 genotypes of safflower. Analysis of variance on different traits showed significant differences among genotypes for number of sub branches at 5% probability level, and at 1% level for other traits indicating genetic variations for different traits in genotypes. Path analysis revealed that the most direct effects were those of the number of heads per plant (0.557). Factor analysis extracted four factors. Factor loading of first three eigen values showed that first three factors explained 97.239%, sum of variance. Rates of first, second and third factors were 35.195%, 31.114/% and 30.193%, respectively. The first, second and third factors were named productivity factor (seed yield and oil yield), sink factor (number of sub branch and number of heads per plant) and fixed capital factor (number of days to 50% flowering, number of days to maturity and plant height), respectively. Cluster analysis divided genotypes into four groups. Group three with 17 genotypes that they had higher number of sub branches, number of head per plant, seed yield and oil yield than general mean. In general, selection indices under rain fed conditions were introduced as seed yield, number of head per plant, number of sub branches, number of days to 50% flowering, number of days to maturity and plant height