Combined Testcross and S1 Selection for the Improvement of Testcross and Inbred Performances

In crops where heterosis is high, one of the limiting factors in the development of commercial single‐cross hybrids is the seed yield of the female parent. To improve simultaneously combining ability with the male parent and the seed yield of the female parent, combined testcross and S 1 selection has already been proposed by several researchers. In this study, after derivation of the theoretical expression of the relative efficiency of combined testcross and S 1 selection, the testcross and S 1 correlated responses are compared with those of direct selection with the same number of replications. The genetic advance using an economic index combining testcross and S 1 performances is also considered. Combined selection can be more efficient than direct selection for the improvement of testcross performance only with a low testcross heritability, a relatively high S 1 heritability, and a high generic correlation between Si and testcross progenies. Combined selection will generally be less efficient in improving testcross performance than direct selection for testcross performance if the same number of plots is used in both methods. Evaluations of selection methods using an economic index that accounts for the cost of seed production demonstrated that combined selection can be more efficient than direct testcross selection. However, the economic weight placed on the S 1 value must be relatively high. With the same total number of replications as for testcross selection, combined selection will generally be more justified with unimproved than with elite material.