Molecular and morphological evaluation of doubled-haploid lines in maize. 2. Comparison with single-seed-descent...

Doubled-haploid (DH) and single-seed-descent (SSD) lines in maize have been compared for quantitatively inherited traits and for RFLP markers. The comparisons of the distributions for agromorphological traits do not allow definite conclusions to be drawn on the similarity of the two reproductive systems. We have used more than 100 RFLP markers to provide a precise description of the parental allele frequency and the recombination fractions. A comparison of two DH populations shows that non-random meiotic reassortment is influenced by differences in the anther culture capacities of the two parental lines. For the DH lines derived from the cross DH5 x DH7, involving two responsive lines in anther culture, the distortion in segregation (P < 0.05) affected less than 20% of the genome with half of the deviations towards each parent. DH lines derived from the cross A188 x DH7, where A188 is a non-responsive line, showed more than twice this level of distortion and an excess of DH7 alleles was found for almost all of the skewed loci. The recombination fractions were homogeneous between the two DH populations for most of the genome. The genome sizes calculated with the DH and the SSD lines derived from the same cross, A188 x DH7, were also similar, which suggests that no selection against recombinant gametes occurs during anther culture. The observed recombination fraction after five meioses (SSD) is on average twice as large as after one meiosis (DH). No difference is observed for recombination fractions greater than 20%. Despite a precise description of the material at the molecular level, it has not been possible to make a definite conclusion as to whether or not the differences in some morphological characters are the consequences of differences in the segregation ratio and/or the recombination frequency. However, the agromorphological evaluation shows a narrow range in differences between the two types of lines and suggests that the use of DH lines is possible in breeding programmes.