Many microbial turnover processes in acidic sandy subtropical soils are still poorly understood. In a 59-day pot and a 189-day laboratory incubation experiment with two West African continuous cereal soils, the effects of 2 mg g -1 root residues were investigated on growth of sorghum seedlings, soil microbial biomass and activity indices, using cowpea, groundnut, pearl millet, maize and sorghum. The effects of root residues were compared with mineral P or mineral P+N treatments and with a non-fertilized control treatment. On the Alfisol (Fada, Burkina Faso), shoot dry mass was always significantly higher than on the Ultisol (Koukombo, Togo). Highest shoot dry mass was observed after application of mineral P+N on the Alfisol and after mineral P alone on the Ultisol. The application of legume root residues led to small and non-significant increases in dry mass production compared to the non-amended control, whereas the application of cereal root residues led to a decline, regardless of their origin (millet, maize or sorghum). Contents of microbial biomass C, microbial biomass N and ergosterol were 75 to 100% higher in the Alfisol than in the Ultisol, while ATP was only 36% higher. Organic amendments increased ergosterol concentrations by up to 145% compared to the control and mineral P application. Microbial biomass C and microbial biomass N increased by up to 50% after application of root residues, but ATP only up to 20%. After application of legume root residues, cumulative CO 2 production was similar in both soils with an average of 370 g CO 2-C g -1 over 189 days. After application of cereal root residues, cumulative CO 2 production was higher in the Alfisol (530 g g -1) than in the Ultisol (445 g g) over 189 days.
