Three successive generations were produced between 2001 and 2003 to assess to what extent genetic variability exists for survival in juvenile oysters Crassostrea gigas. For each generation, two groups were selected for their high ("R" for resistant) and low ("S" for susceptible) survival. Significant differences in mortality were observed during the first year in Riviere d'Auray (RA) (Brittany-France). However, low and similar mortality were observed for both groups, with no significant difference between groups in the first and second generation in this site, where the first summer is the critical, sensitive period with a clear age effect. In Baie des Veys (BDV) (Normandie-France), where mortalities usually mainly affect adults, mortality between groups, tested in the third generation, was low and similar during the first year, but higher mortality was observed during the second summer period, showing significant difference between groups. Additionally, both groups were also retained at our nursery (Bouin) during the first year to preserve them from field mortality pressure. These oysters were then transferred to RA and BDV during the spring. Significant differences of mortality were found between the "R" and "S" groups after the summer period. Mortality rates were lower than those observed when spat were deployed the first year in the field in RA and the opposite result was found in BDV. These findings should allow the oyster commercial industry to optimize rearing strategy according to the environment and the age of oysters in order to minimize mortality.