The combining ability of eight wheat varieties was evaluated according to the fourth model of the Griffing's diallelic methodology. Studies on genetic dissimilarity based on microsatellite markers and genetic distance among genotypes from pedigree data were also performed. The experiment was carried out in a randomized block design with two replications. Eight traits were evaluated in the diallel. Genotypes were grouped according to the UPGMA and Tocher methods. Genetic variability among genotypes was evident. Varieties 'CD 108', 'CD 0542' and 'CD 104' were those who showed high values for general combining ability in several traits. Since the effects of specific combining ability were more important in those particularly heterozygous combinations obtained from varieties allocated in different clusters, field and molecular results coincided in a certain way. There was no a good coincidence between the dendrogram based on parentage coefficient and the one based on microssatellite markers. The very small association between standards of grouping was probably related to the intrinsic properties of each way of estimating the genetic distance, which can modify the interpretation and the distribution of the genetic variability in the evaluated genotypes.
Triticum aestivum; parentage coefficient; diallel crosses; dissimilarity grouping
