Detection of QTLs is usually performed on the basis of a F2 segregation strategy involving two characterized parents. Such approach is difficult to apply in long generation-time and wild species like forest trees. Thus, we develop another method for detection of QTLA in several F1 families connected in a mating design. This original approach is based on the analysis of relationships between parent grootype and progeny performance. Quantitative traits are evaluated at the family level, which provides better evaluation than F2 individual values. The familial performances related to the familial allelic frequencies of the markers deduced from the respective parental genotypes. Several markers correlated with quantitative traits among families can be arm then thus identified; moreover, they are not specific of a peculiar parent. We expect that this gametic disequilibrium between QTAs and alleles at some of these marker loci results from linkage. Therefore, the QTI. detection in such mating design is assumed to be suitable for marker-aided selection. This strategy has been applied on a factorial mating design involving 12 European and 12. Japanese larches. Progeny performance for growth, architectural traits and wood specific gravity has been assessed for 11 years determined using RAPD markers on DNA needie extracts and on megagametophytes in a field trial. Genotypes of parents were to identify heterozygous individuals. Two hundred and ten primers were tested on 24 individuals. Nine markers were correlated to one or two of the investigated traits. Genotypes of individuals of appropriate families selected on the basis of parental genotypes will be analysed at the retained marker loci, and linked to their individual performance in order to precise the relationships between these markers and QTLs