In France, as in other countries, there is a need to improve knowledge of greenhouse gas and ammonia emissions from livestock production with various management options. The aim of this study was to quantify the NH3, N2O, CO2 and CH4 emissions from two pig solid manure heaps, turned and not turned, in order to examine the effect of turning on gaseous emissions and to obtain additional information about emissions during composting in commercial conditions. The manure produced from the fattening of 72 pigs on straw bedding was divided between two identical heaps and stored outside on a concrete area for three months during autumn 2004. One heap was turned at day 0, 11 and 27. Gaseous emissions (NH3, N2O, CH4, CO2, H2O) of both heaps, covered for the purpose by ventilated greenhouses, were continuously measured over four periods (57 days in total). The tracer method, using SF6, was used to measure the ventilation rate in the greenhouses. Concentrations inside and outside the greenhouses (to have a differential) were measured by photoacoustic infrared absorption spectrometry with a gas analyser coupled to a multipoint sampler and doser. Emissions were calculated by combining various flows and differential concentrations and interpolated between the different measuring periods. Results show that emission kinetics were very similar for the two heaps, except after turning, when emissions are given a new impetus. On average, the measured gas emissions for unturned and turned heaps represent a nitrogen loss (mainly in the form of NH3) of 10.5% and 9% of the initial nitrogen, and a carbon loss (mainly in the form of CO2) of 24 et 28% of the initial carbon. For both heaps, N2O is the primary greenhouse gas emitted (3% of the initial nitrogen) and not CO2 or CH4. Considering the uncertainty on the measurements and the representativity of the experiment, these results did not show any clear difference between the two heaps. They underlined the relationship between gaseous emissi