Factors influencing rates of C and N mineralization of soil and plant materials, and the reliability of different procedures for estimating microbial biomass, were measured in a soil (Typic Dystrochrept) that had been restored under grazed pasture in a temperate environment for 10-11 years after 20 cm of the original topsoil had been removed by stripping.Rates of net N mineralization were appreciably lower, but C02-C production higher, in the stripped than in the unstripped soil. These activities were not influenced directly by levels of soil mineral-N, but they were influenced by differences in plant composition. Herbage and litter, and roots, from the stripped plots were generally mineralized more readily to CO2-C, but more slowly to net mineral-N, than were the corresponding materials from the unstripped plots. Rates of mineralization of herbage and litter, or roots, were mainly indistinguishable in stripped and unstripped soil, whereas rates of mineralization of all standing dead material were lower in stripped soil.