Estimation of heterosis and combining abilities of U.S. winter wheat germplasm for hybrid development in Texas

Abstract Hybrid wheat ( Triticum aestivum L.) offers promises to break the yield stagnation in global wheat productivity as a result of heterosis. But, for this promise to be realized, the level of heterosis must be adequate. To test this, elite winter wheat lines from the breeding programs of the University of Nebraska–Lincoln (UNL) and Texas A&M University (TAMU) were crossed in a 25‐by‐25 full‐diallel design using a chemical hybridizing agent (CHA) to produce experimental hybrids. These hybrids were planted in a modified augmented design with commercial checks and parents at McGregor, TX, in 2016 ( n = 612) and Greenville and Bushland, TX, in 2017 ( n = 470) to evaluate for yield heterosis and combining ability. A subset of hybrids ( n = 333) were repeated between years. The effect of field heterogeneity in grain yield was corrected by spatial modelling in ASReml‐R. Commercial heterosis ranged from −78.3 to 20.4% in 2016 and −32.9 to 6.2% in 2017. High‐parent heterosis (HPH) ranged from −70.4 to 54.3% in 2016 and −26.9 to 29.2% in 2017. General combining ability (GCA) variance was significantly higher than zero, whereas specific combining ability (SCA) variance was not. Significant maternal effect was identified in some crosses tested by computing reciprocal effects between crosses and estimating variances. This indicates that most of the heterosis is due to additive rather than dominance effect. These results suggest exploitation of GCA for higher yield while underscoring the need for development of heterotic pools to maximize SCA and dominance effects.