ASSOCIATION PATTERN AND REGRESSION ANALYSIS FOR YIELD AND ITS ATTRIBUTING TRAITS IN SAFFLOWER (Carthamus tinctorius L.)

Abstract

The present study was conducted to determine the interrelation among yield and its related traits in safflower using a RIL population derived from the cross between CO-1 and EC-523368-2. The RILs were phenotyped for seven traits, days to 50% lowering, days to maturity, plant height, number of primary capitula/plant, number of seeds/primary capitulum, 100 seed weight
and seed yield. Trait mean in the population was observed as 82.08 days for days to 50% flowering, 112.15 days for days to maturity, 75.62 cm for plant height, 6.71 primary capitula/ plant, 28.96 seeds/ primary capitulum, 3.29 g for 100 seed weight. Mean
of seed yield was observed as 9.44 ranging from 1.99 to 34.38 g. Correlation analysis revealed the strong positive correlation of
seed yield with number of seeds/ primary capitulum, number of primary capitula/ plant, plant height and 100 seed weight. The
stepwise regression analysis explained 47% of variation for seed yield (coefficient of determination R2=0.47) and adj. R2 as 0.45
and developed a best fit linear regression equation as, SY = -9.55 + 0.06PH + 0.77NPCP + 0.21NSP + 1.06HSW. Based on Mallows’ Cp (low- 4.18), R2 (high- 87.45) and adj. R2 (high-87.01), the best fit regression model from all possible models was PH + NPCP + NSP + HSW. Selection for these traits may help to improve the seed yield potential in safflower.