Causes of agronomic differences between synthetics developed by the random and convergent cross methods

The theoretical utility of two alternative methods, random (r) and convergent (c) cross methods, of producing maize synthetics has previously been ascertained in two different genetic backgrounds (EPS20 derived from eight Reid inbred lines with origin from the US Corn Belt population “Reid”, and EPS21 formed from eight non-Reid inbreds). However, the agronomical consequences of using one or another methodology have not been tested. The objectives of the present study were to determine, in two genetic backgrounds, whether synthetics developed by the random and convergent cross methods differed in agronomic performance and to investigate whether some allelic changes previously observed by Butrón et al. [Butrón, A., Tarrío, R., Revilla, P., Malvar, R.A., Ordás, A., 2003. Molecular evaluation of two methods for developing maize synthetic varieties. Mol. Breed. 12, 329-333.] could be directly implicated in those differences. The synthetics and the diallel crosses among them, testcrosses of EPS20c and EPS20r to their Reid parental inbreds and testcrosses of EPS21c and EPS21r to their non-Reid parental inbreds were evaluated for grain yield in three trials in 2004 and 2005. Our results suggest that directional selection for germination, which occurs during the process of formation of synthetics using the random method (but absent with the convergent cross method) was responsible for agronomic and genetic differences between synthetics obtained by alternative methods from the same set of inbreds. Although selection for germination increased the yield performance of the synthetic obtained from the Reid inbreds, in a more heterogeneous genetic background, natural selection against non-competitive inbred lines at germination would be responsible for an important reduction of variability that would reduce yield.