Captive-reared animals released as part of reinforcement programmes are considered less likely to survive than their wild conspecifics. One of the possible causes for this is reduced digestion efficiency. We studied adaptation of the digestive system in Mallard, a species with high adaptability to its environment. Body condition and digestive organs were compared between three groups: captive-reared Mallards remaining in a game farm, captive-reared Mallards released into the wild as juveniles and wild Mallards. We also assessed difference in diet between released birds and wild birds, and conducted a one-year survival analysis of captive-reared birds released in a hunting-free area. Released Mallards had a smaller gizzard than wild birds, but there was no difference between captive-reared and wild Mallards in the size of others organs in the gastro-intestinal tract. Body condition of captive-reared Mallards was poorer than wild ones, and this was only partially improved after release. Survival probability of captive-reared Mallards was low, compared to documented survival of wild Mallards. In particular, high mortality occurred when additional food provisioning was stopped and during harsh winter periods. We argue that in spite of a viable digestive system, pre-release conditions experienced by captivereared ducks impeded restocking success. In the context of massive releases of this species for hunting purposes (several million birds per year in Europe), low survival due to a combination of high hunting pressure and poor condition could limit the number of farmed birds surviving to breed and thus limit introgression of "captive genes" into the wild population.