Abstract Little is known about the effect of addition of plant residues from which water-extractable organic C (WEOC) was partially removed on soil microbial activity and growth. Two incubation experiments were conducted in a sandy clay loam amended with barley (Hordeum vulgare L.) residues (ground to 0.5–1 mm) from which WEOC was removed by extraction (Experiment 1) or leaching (Experiment 2). For the first experiment, the residues were extracted 5 times to maximise WEOC removal. There were three treatments: unamended control soil, soil amended with original residue or with extracted residue. Cumulative respiration increased rapidly within the first 5 days with the original residues, whereas it started more slowly with extracted residues. By the end of the experiment (day 18) compared to the original residues, cumulative respiration was 18% lower with extracted residues, but the microbial biomass C (MBC) concentration was 27% greater. For the second experiment, the residue was leached 1 to 8 times to reduce WEOC concentrations by 29 and 80%, respectively. There were four treatments: unamended control, soil amended with original residue, residue leached once and eight times. Cumulative respiration was reduced by up to 21% by partial removal of WEOC compared to the original residues in the first 10 days, but by day 28 (end of experiment) it was greatest with residues leached 8 times where it was 5% greater than with the original residues. This treatment also had a 16% higher MBC concentration on day 28 than soil with the original residue. It is concluded that nearly complete removal of WEOC reduced soil respiration compared to the original residues in the first week after residue addition. However, when removal is incomplete and a small proportion of WEOC (20%) remains in the residues, respiration may increase later above that of the original residues.