120 waterfowl of 3 types were raised at the station experimentale des
palmipedes a foie gras in Artigueres as follows: 40 Landaise geese, 20
males and 20 females, 40 mule ducks, males only, 40 muscovy ducks, ma
les. The feeding program was specific to each type of bird. The purpos
e of such an approach was to adapt birds in the best way before crammi
ng. However, numerous parameters (choice of corn, cramming room, mater
ial and operator) have been standardized. Cramming lasted for 13 d At
the end of this period, each type of bird presented a spectacular fatt
ening, ranging from +41% for muscovy ducks to +53% for geese and +54%
for mule ducks. Furthermore, ''foie gras'' production was also very di
fferent between each type or waterfowl since it ranged from 560 to 793
g for the muscovy ducks and geese, respectively The performance of mu
le ducks was intermediate (702 g) and significantly different from the
other waterfowl. The technical quality of ''foie gras': appreciated b
y the amount of fat loss during thermic processing, showed the same cl
assification: 21.2% for geese, 43.9% for mule ducks and 55.6% for musc
ovy ducks. Muscovy ducks present the best yield for breast muscles; ge
ese are the least productive for this parameter, but they present the
best yield for thigh muscles. It appears that starting the cramming tr
eatment at 13 weeks for geese does not correspond to the total maturit
y of these birds. Depending on the purpose of each one, all these wate
rfowl can be used, but it seems that there is by now an overall advant
age to using mule ducks for cramming.