Combining Ability of 54 Sunflower Paternal Lines Crossed with Three CMS Testers: Additive Control of Oil Content and Seed Weight and the Choice of Donor Lines

For sunflower (Helianthus annuus L.) hybrid breeding, we evaluated 54 paternal lines crossed with three cytoplasmic male-sterile testers. The 162 hybrids were grown for two years at one location and analysed with a plot-level mixed model that separates genotype × year terms. Baker’s ratio was 0.958 for oil content, 0.951 for 1000-seed weight and 0.933 for plant height, and the heritability of the hybrid mean 0.90 and 0.91 for the first two. Without the two confectionery lines, Baker’s ratio for these traits remained 0.877 and 0.907 and heritability 0.72 and 0.83. Paternal general combining ability (GCA) for oil content, 1000-seed weight, plant height and flowering time agreed between the two seasons (concordance 0.71–0.85). For seed yield, additive control was not established (Baker’s ratio 0.440, 95% interval 0.00–0.82); plot error was 3.6 times the genetic variance and paternal GCA was weakly repeatable (r = 0.24). VK101A raised oil content, VK934A 1000-seed weight and plant height, and VA761A lowered plant height. A composite index identified LI18, LI41, LI6, LI53 and LI19 as candidate donors. Paternal lines can be selected on GCA for these three traits; seed yield and the ranking of donors that depends on this yield require testing in more environments.