Screening of sunflower populations for seed yield and its components through step-wise regression analysis

摘要

Improving achene yield is the principal breeding objective for sunflower improvement and its commercial acceptance. Therefore, selection was practiced in six sunflower populations for seed yield. Step-wise regression analysis revealed two variable model including head diameter and achene weight for the improvement of sunflower populations 2 and 3 accounting 48.7 and 73.0% variability in seed yield per plant, respectively. Sunflower population 5 accounted minimum variability 42.6% for improvement by including achene weight, leaves per plant and distance from head to soil surface. However, populations 1 and 4 accounted maximum variability of 73.0 and 74.3%, respectively for seed yield per plant. Four variable (head diameter, achene weight, bird damage and stem diameter) and five variable models (achene weight, head diameter, plant height, leaves per plant and internodal length) were best fitted for sunflower populations 4 and 1, respectively. From the study it is evident that selection of head diameter and 100-achene weight may be effective for seed yield improvement in sunflower.