Recurrent Selection for Seed Size in Soybean: I. Response to Selection in Replicate Populations

Abstract

Seed size is an important component of soybean [Glycine max (L.) Merr.) yield. The objectives of this experiment were to investigate the response to recurrent mass selection for increased seed size on male‐sterile (MS) plants, examine the indirect response on seed size and seed yield of derived male‐fertile (MF) plants, and evaluate the repeatability of selection response across replicate populations. Plant germplasm was developed from the intermating population N79‐1500, which segregates for male sterility. The base population was divided into three replicate subpopulations and MS plants were selected for large seed size in each separately. After four cycles of selection in two of the subpopulations and three cycles in the third, composites of the parents of each cycle were tested at four locations in North Carolina for 2 yr. There was a significant (P ≤ 0.01) increase in both MS and MF seed size for all subpopulations; the rate of increase in MS seed size was approximately twice that of MF seeds. The increase in mean MF seed size was associated with an increase in the frequency of seeds in the larger size classes and a decrease among the smaller classes. Seed yield increased linearly (P ≤ 0.01) for two of the subpopulations at rates of 46 ± 18 and 78 ± 27 kg ha−1 cycle−1. The third showed a nonsignificant increase (17 ± 19 kg ha−1 cycle−1). Recurrent mass selection for increased seed size facilitated by genetic male sterility may be an effective adjunct in a conventional breeding program to increase the yield potential of populations yet require only slightly more resources.