A technique for measuring net rates of mineralization under field conditions is described. Soil cores were incubated in the field in sealed containers with acetylene to inhibit nitrification and thereby minimize losses of N through denitrification. Mineralization was estimated as the difference between the mineral N content after a 14-d incubation and that determined from soil samples taken at the start of incubation. Mineralization in the spring and summer in unfertilized plots in the field amounted to 90 and 70 kg N ha-1 in S.E. England under grass and grass/clover swards, respectively, and 40 kg N ha-1 under a grass sward in S.W. England. Daily rates of mineralization ranged from 0.02 to 1.90 kg N ha-1, with peak values related to re-wetting of the soil after dry weather. Laboratory incubation of soil showed that neither the low concentration of acetylene (2% v/v) adopted for field incubation, nor the accumulation of mineral N during incubation was likely to affect the total measurement, but that frequent and regular soil sampling was necessary to minimize the effects of changes in soil water content. Estimates for mineralization over the whole growing season (180 d) were obtained for two years from extrapolation of the early season field measurements and were, on average, 50% higher than predictions based on a chemical extraction index of potentially mineralizable N. © 1996 Kluwer Academic Publishers.