This study investigated the effects of various stunning methods on the quality of French 'foie gras' in ducks and geese. The ducks (n = 30 per group) were stunned with one of the following techniques: electrical stunning in a water bath (50 Hz AC, 130 mA, 4 s), head-only electrical stunning (50 Hz AC, 600 mA, 4 s), mechanical stunning (captive bolt) and controlled atmosphere stunning (CAS: phase 1, CO2 (40%)-O2 (30%)-N2 (30%), for 2 min followed by phase 2, CO2 >85% in air, for 2 min). The same methods (except head-only stunning which was not applied) were used in geese (n = 40 per group). The weight of the liver at slaughter was not affected by the stunning techniques, neither was its colour (L*, a*, b* coordinates), despite the differences in bleeding rate between the stunning treatments. The loss of fat during the cooking of canned fatty liver did not depend upon stunning treatment. Some appearance defects of raw fatty livers were significantly affected by the treatment: mechanical and head-only stunning induced higher incidence of petechial haemorrhages in duck liver, whereas in geese, incidence of superficial haemorrhages was significantly higher after CAS and water-bath stunning. The calculation of an overall score based on the incidence and severity of the different appearance defects observed in the present experiment showed that CAS was associated with the least favourable position in ducks as well as in geese, compared to the other treatments. The commercial grading of ducks and geese fatty livers, carried out by an expert from the industry, clearly showed the detrimental effect of CAS on the commercial value of raw livers. CAS, under the conditions applied in the present work cannot be recommended, because of its drastic effect on liver quality. The underlying mechanisms deserve further investigations.