Plants significantly affect rates of carbon (C) turnover in soils, both because they are sources of carbon through exudation in the rhizosphere and litter-fall, and because rhizosphere microbes stimulated by roots also metabolize native soil carbon. Different plant species affect these components of soil carbon turnover in different ways, but the quantitative information on this is lacking for different ecosystems and soil-plant combinations. To compare the effects of grassland and forest plant species on the components of rhizosphere respiration in different soils, we grew ryegrass ( Lolium perenne) and radiata pine ( Pinus radiata D. Don) in two silt loam soils in pots in a glasshouse, and in seven samplings over 45 weeks measured total ( Rtotal), root ( Rroot) and root-free soil respiration ( Rrfs), the latter from respiration in unplanted controls. We calculated rhizosphere respiration ( Rrhizo), defined here as the net of that fuelled by native soil C and root-derived C, from Rtotal less Rroot+ Rrfs. We also measured plant growth and total, water-soluble and microbial biomass C in the soils at each sampling. Results showed that Rrfs decreased over the experimental period in both soils. Under ryegrass, Rroot, Rrhizo and Rtotal increased up to 14 weeks after planting and then stabilized, whereas under radiata pine, they continued to increase throughout the experiment. By the end of the experiment, the Rroot, Rrhizo and Rrfs components accounted for 49–58, 31–50 and 1–11% of soil total respiration under ryegrass, respectively, and 43–66, 29–53 and 1–5% under radiata pine. The greater Rroot, Rrhizo and Rtotal values under radiata pine were related to greater root biomass and root-derived organic C, and enhanced microbial mineralization of native soil organic C. [ABSTRACT FROM AUTHOR]