The effect of incorporating cattle slurry in soil, either by mixing or by simulated injection into a hollow in soil, on the ryegrass uptake of total N and (NH4+)-N-15-N was determined in three soils of different textrue. The N accumulation in Italian ryegrass (Lolium multiflorum L.) from slurry N and from an equivalent amount of NH4+-N in ((NH4)-N-15) SO4 (control) was measured during 6 months of growth in pots. After this period the total recovery of labelled N in the top soil plus herbage was similar in the slurry and the control treatments. This indicated that gaseous losses from slurry NH4+-N were insignificant. Consequently, the availability of slurry N to plants was mainly influenced by the mineralization-immobilization processes. The apparent utilization of slurry NH4+-N mixed into soil was 7%, 14% and 24% lower than the utilization of (NH4)(2)SO4-N in a sand soil, a sandy loam soil and a loam soil, respectively. Thus, the net immobilization of N due to slurry application increased with increasing soil clay content, whereas the recovery in plants of N-15-labelled NH4+-N from slurry was similar on the three soils. A parallel incubation experiment showed that the immobilization of slurry N occurred within the first week after slurry application. The incorporation of slurry N by simulated injection increased the plant uptake of both total and labelled N compared to mixing the slurry into the soil. The apparent utilization of injected slurry NH4+-N was 7% higher, 8% lower and 4% higher than the utilization of (NH4)(2)SO4-N in the sand, the sandy loam and the loam soil, respectively. It is concluded that the spatial distribution of slurry in soil influenced the net mineralization of N to the same degree as did the soil type.