The effects of 5 years of continuous grass/clover (Cont grass/clover) or grass (Cont grass) pasture or 5 years of annual grass under conventional (Ann grass CT) or zero tillage (Ann grass ZT) were compared with that of 5 years of continuous barley (LT arable) on a site which had previously been under arable crops for 11 years. For added comparison, a long-term grass/clover pasture site (LT past) nearby was also sampled. Soil organic C (Corg) content followed the order LT arable=Ann grass CT<Ann grass ZT<Cont grass=Cont grass/clover<LTpast. Trends with treatment for microbial biomass C (Cmic), basal respiration, flourescein diacetate (FDA) hydrolytic activity, arginine ammonification rate and the activities of dehydrogenase, protease, histidase, acid phosphatase and arylsulphatase enzymes were broadly similar to those for Corg. For Cmic, FDA hydrolysis, arginine ammonification and the activities of histidase, acid phosphatase and arylsulphatase, the percentage increase caused by 5 years of continuous pasture (in comparison with LT arable) was 100–180%, which was considerably greater than that for organic C (i.e. 60%). The microbial metabolic quotient (qCO2) was higher for the two treatments which were mouldboard ploughed annually (LT arable and Ann grass CT) than for the undisturbed sites. At the undisturbed sites, Corg declined markedly with depth (0–15 cm) and there was a similar stratification in the size and activity of Cmic and enzyme activity. The microbial quotient (Cmic/Corg) declined with depth whilst qCO2 tended to increase, reflecting a decrease in the proportion of readily available substrate with depth.