The annual carbon dioxide (CO 2 ), nitrous oxide (N 2 O) and methane (CH 4 ) dynamics were measured with static chambers on two organic agricultural soils with different soil characteristics. Site 1 had a peat layer of 30cm, with an organic matter (OM) content of 74% in the top 20cm. Site 2 had a peat layer of 70cm but an OM content of only 40% in the top 20cm. On both sites there were plots under barley and grass and also plots where the vegetation was removed. All soils were net sources of CO 2 and N 2 O, but they consumed atmospheric CH 4 . Soils under barley had higher net CO 2 emissions (830g CO 2 –C m −2 yr −1 ) and N 2 O emissions (848mg N 2 O–N m −2 yr −1 ) than those under grass (395g CO 2 –C m −3 yr −1 and 275mg N 2 O–N m −2 yr −1 ). Bare soils had the highest N 2 O emissions, mean 2350mg N 2 O–N m −2 yr −1 . The mean CH 4 uptake rate from vegetated soils was 100mg CH 4 –C m −3 yr −1 and from bare soils 55mg CH 4 –C m −2 yr −1 . The net CO 2 emissions were higher from Site 2, which had a high peat bulk density and a low OM content derived from the addition of mineral soil to the peat during the cultivation history of that site. Despite the differences in soil characteristics, the mean N 2 O emissions were similar from vegetated peat soils from both sites. However, bare soils from Site 2 with mineral soil addition had N 2 O emissions of 2–9 times greater than those from Site 1. Site 1 consumed atmospheric CH 4 at a higher rate than Site 2 with additional mineral soil. N 2 O emissions during winter were an important component of the N 2 O budget even though they varied greatly, ranging from 2 to 99% (mean 26%) of the annual emission.