[en] Five successive batches of fattening pigs were raised, each during a four month period, on a totally concrete slatted floor in one experimental room and on straw based deep litter in another. The rooms were automatically ventilated to maintain a constant ambient temperature. Available floor space was of 0.75 in m(2) per pig kept on the slatted floor and 1.20 m(2) per pig kept on the deep litter. With this last system, about 46 kg of straw were supplied per pig throughout a fattening period. The slurry pit was emptied and the litter removed after each batch. Once a month, the emissions of ammonia (NH3) nitrous oxide (N2O), methane (CH4), carbon dioxide (CO2) and water vapour (H2O) were measured continuously for 6 consecutive days by infra-red. photoacoustic detection. The performance of the animals was not significantly different according to the floor type. Gaseous emissions from pigs raised on the slatted floor and on the deep litter were, respectively, 6.2 and 13.1 g per pig per day for NH3, 0.54 and 1.11 g per pig per day for N2O, 16.3 and 16.0 g per pig per day for CH4, 1.74 and 1.97 kg per pig per day for CO2 and 2.48 and 3.70 kg per pig per day for H2O. Except for the CH4 emissions, all the differences were significant (P<0.001). Thus, pig fattening on deep litter releases nearly 20% more greenhouse gases than on slatted floor, with 2.64 and 2.24 kg of CO2 equivalents, respectively (P<0.001). Whatever the floor type, emissions increased from the beginning to the end of the fattening periods by about 5 times for NH3, 4 times for N2O, 3 times for CH4 and 2 times for CO2 and H2O. Correlation coefficients between CO2-emissions and H2O, NH3 and CH4 emissions were, on average for both floor types, 0.82, 0.77 and 0.74, respectively. Although rearing pigs on straw generally has a good brand image for the consumer, this rearing system produces more pollutant gases than keeping pigs on slatted floors. (c) 2007 Elsevier B.V. All rights reserved.