The microbial immobilization and subsequent transformation of SO4(2-)-S were monitored in six soils following amendment with S-35-labelled SO4(2-)-S (20 mug S g-1 soil) with or without glucose (2000 mug C g-1 soil). In the soils receiving glucose, 21-34% of the added SO4(2-)-S-35 was immobilized within 3 days whereas without addition of glucose, only 0.5-13% was immobilized. Similar proportions of the added SO4(2-)-S-35 were converted into microbial biomass (biomass-S-35) over the same time. Over 127 days, both amendments (with or without glucose) had little effect on soil SO4(2-)-S and extractable-S. There was no measurable effect of the glucose-free amendment on soil biomass-S. However, in soils receiving glucose, total biomass-S increased markedly over the first 3 days (11-24 mug S g-1 soil), whereas in these soils, the amounts of labelled biomass-S-35 formed over the same period (3.3-8.7 mug S g-1 soil) were much lower. The additional increase in total biomass-S was due to a priming effect of the glucose amendment on the decomposition of soil organic-S. Between 3 and 10 days, total biomass-S in soils receiving glucose declined to concentrations similar to those in soils receiving the glucose-free amendment. The S (both labelled and unlabelled) lost from this decrease in total biomass-S was converted into soil organic-S since it was undetectable in the SO4(2-)-S pool. The data suggest that the turnover of microbial biomass is the primary process by which S is transformed into soil organic matter.