Genetic variation and time trends for growth and stem quality traits were evaluated in jack pine (Pinus banksiana Lamb.) second-generation progeny tests. The study was based on a large dataset, involving 275 half-sib families and 17 field trials, planted across New Brunswick, Canada. Growth was indicated by cumulative tree height (HT) measured at trial ages from 5 to 20 years, and stem quality was evaluated on stem straightness (SST) and stem forking (SF). Results indicated that both HT and SST were moderately genetically controlled, with mean individual heritability estimates of 0.14 (range: 0.03-0.30) and 0.16 (range: 0.10-0.25), respectively, which were lower than those estimates from the first-generation progeny tests. Heritability estimates for HT increased from trial ages of 5 to 10 years and then remained stable, whereas for SST, heritability estimates showed a "⌳" shape with the highest value occurring at trial age of 15 years. HT measurements at trial ages of 5 to 10 years were highly predictive of HT growth at 15 to 20 years, and similarly, SST assessment at 10 years correlated well with assessment at 20 years. Estimates of genetic correlation between HT and SST were positively and moderately strong, which contrasted with the moderately negative genetic correlation between the two traits in the first-generation progeny. Heritability estimates for SF were generally low (mean h i 2 = 0.06) and remained constant over time. SF was essentially not genetically correlated with HT or SST. Family-by-site interactions were weak for HT and SST over the trial ages but were strong for SF at trial age of 10 years or older. The potential impact of selection on genetic architecture of growth and stem quality traits in advanced generation jack pine breeding populations and on breeding and selection strategies was discussed.Résumé : La variabilité génétique et ses tendances temporelles pour les caractères de croissance et de qualité de la tige ont évaluées au sein de tests de descendances de seconde génération de pin gris (Pinus banksiana Lamb.). L'étude s'est appuyée sur un important jeu de données impliquant 275 descendances uniparentales et 17 tests au champ établis à travers le NouveauBrunswick au Canada. La croissance a été déterminée comme la hauteur cumulative des arbres (HT) mesurée de 5 à 20 ans après la plantation, alors que la qualité de la tige a été évaluée à partir de la rectitude du fût (SST) et de la présence de fourches (SF). Les résultats ont indiqué qu'à la fois HT et SST étaient modérément sous contrôle génétique avec des valeurs moyennes d'héritabilité individuelle respectives de 0,14 (écart : 0,03-0,30) et de 0,16 (écart : 0,10-0,25), qui étaient plus faibles que les valeurs obtenues à partir des tests de descendances de première génération. Les valeurs d'héritabilité de HT augmentaient entre 5 et 10 ans après la plantation, et demeuraient stables par la suite alors que dans le cas de SST, ces valeurs prenaient la forme « ⌳ » avec la plus forte valeur obtenue 15 ans après la plantation. Les mesures ...