Field experiments were conducted to test a new approach for estimating crop N uptake from organic inputs. Soils were pre-labelled by applying N-15 fertiliser to soybean [Glycine max (L.) Merr] and common bean [Phaseolus vulgaris L.] crops. Additional N-14 plots which received unlabelled fertiliser were also established in the same way. The above-ground biomass from all four plots was harvested, stored and the plots left to over-winter. In the following summer N-15-labelled residues were added to the unlabelled soils and unlabelled residues were added to the 15N-labelled soils at a rate of 150 kg N ha(-1). All plots were cultivated and sown with maize (Zea mays L.). Control plots that did not receive residue application or any additional fertiliser N were also set up. The plots were harvested in late autumn. Maize derived 37 kg N ha(-1) and 31 kg N ha(-1) from the added soybean residues, estimated using the direct and indirect approach, respectively, in plots previously sown to soybean. N derived from the added common bean residues was estimated as 26 kg N ha(-1) and 24 kg N ha(-1) from the direct and indirect methods, respectively. In the plots previously sown to common bean, N derived from added soybean residue was 32 kg N ha(-1) using the direct method and 33 kg N ha(-1) using the indirect method. N derived from common bean residues was 22 kg N ha(-1) and 21 kg N ha(-1) estimated using the direct and indirect approaches, respectively. It was concluded that the modified indirect technique allows a reasonable estimate of N derived from residues and that this will enable further experiments to be conducted in which N derived from more complex matrices, such as manure or sewage sludge, can be determined.