General and specific combining ability of <i>in vitro</i> doubled haploid maize lines in the field

The breeding of hybrid maize now has a history of over 100 years. In 1908, George H. Shull was the first to report on the high yields, great uniformity and homogeneity of hybrids derived from a cross between two inbred lines. Following this discovery, consistent self-fertilisation over a period of six to eight generations was found to be an extremely efficient method for developing maize lines. From the mid-1970s, however, with the elaboration of the monoploid (<i>in vivo</i>) and microspore culture (<i>in vitro</i>) techniques, it became possible to develop homozygous lines within a year.With the help of an efficient plant regeneration system based on anther culture, large numbers of doubled haploid (DH) lines can be produced. In the course of the experiments the seed of DH plants selected over several years was multiplied and crossed with Martonvásár testers, after which the hybrids were included in field performance trials in three consecutive years (2005–2007). The aim was to determine whether the field performance of hybrids developed in this way equalled the mean yield of standards with commercial value. The data also made it possible to calculate the general (GCA) and specific (SCA) combining ability of the parental lines, indicating the usefulness of the parental components in hybrid combinations and expressing the extent to which a given line contributes to yield surpluses in its progeny.A total of 52 maize hybrids were evaluated in the experiments in terms of yield and grain moisture content at harvest. The combinations, resulting from crosses between 12 DH lines, one control line developed by conventional inbreeding and four testers, were found to include hybrids capable of equalling the performance of the standards, and four DH lines were identified as improving the yield level of their progeny. As the experiment was carried out on a very small number of genotypes, the results are extremely promising and suggest that, if the range of genotypes used to develop DH lines is broadened and the sample number is increased, it will be possible in the future to find maize hybrids, developed with <i>in vitro</i> DH parental components, that surpass the performance of commercial hybrids.