The French project "Morest", coordinated by IFREMER, aims to a better understanding of the causes of summer mortalities in juvenile oysters by a multidisciplinary approach. Within this program, a first generation (G1) was bred in 2001 to assess to what extent genetic variability exists for this trait in French populations. 17 half-sib families (HSF) including 44 full-sib families (FSF) were obtained in the G1, and each FSF was reared in three sites along the French coast during the summer period. Significant differences in survival were observed among HSF, clearly indicating a genetic basis for survival. Estimation of heritability for survival was very high: 0.81  0.29. In 2002, 3 "good" HSF and 3 "bad" HSF were selected on their survival performances to constitute a second generation (G2) by divergent selection. Survival of the G2 was recorded in the same 3 sites according to the same protocol as in 2001, and additionally, the G2 was also monitored in several laboratory experiments. Significant differences in survival were found between the offspring of the "high" and "low" selected groups for all experiments, and the realized heritabilities of the survival in the field were 0.75 for the high group and 0.71 for the low group. These results confirm those observed for the G1, i.e. a high heritability for survival in juvenile cupped oysters, indicating that selective breeding programs could efficiently improve survival of juvenile oysters.