||The most significant genetic improvement for the production of Pacific oyster (Crassostrea gigas) in France to date has been through the breeding of triploids, especially since the development of tetraploids. Triploid oysters commonly resist summer mortality better than diploids. This is most probably because reproductive allocation is so important in this phenomenon. Quantitative genetics studies strongly suggest that a significant gain in resistance to summer mortality, or other traits of interest, could be obtained by selective breeding. Practical difficulties however, and the high cost of breeding large numbers of families under common conditions over many generations remain major constraints to family-based selective breeding in oysters. In the last decade, family-based selective breeding programs have been initiated in the U.S.A., Australia and New Zealand, mainly to improve growth or yield. In France, where most of the spat used for oyster production is collected from wild, no large scale selective breeding program has yet been started, notably because such an initiative requires long term investment to ensure correct management of the selected broodstock (i.e. maximizing genetic variability and avoiding inbreeding). The mixed-family approach, using multiplexed-microsatellite markers to trace parentage, can considerably ease selective breeding, as demonstrated in several fish species. However, high frequencies of null alleles and high variance in reproductive success at early stages have been reported in oysters, leading to potential difficulties applying this approach to the improvement of summer mortality resistance. The combination of selective breeding and polyploidization is a promising but slow and complex process. There is a need for innovative breeding programs and long term investment.