Nitrogen fertiliser recovery of a wheat (Triticum aestivum) cultivar (Seri 82) and residual N remaining in soil after harvest were measured using N-15 labelled fertilisers during four years of field experiments from 1994/95 to 1997/98. Urea, as N-fertiliser, was applied in two splits, a commonly used practice in the region, 1/3 at emergence and the remaining two thirds at tillering, at rates of 0, 80, 160 and 240 kg N ha(-1). Recovery of fertiliser N by the wheat crop was 50-60%, indicating that a significant amount of N fertiliser (40-50%) applied to wheat remained unused in the soil or was lost. However, over 90% of the fertiliser-N applied could be accounted for in the crop or soil after harvest. Of the residual N-fertiliser remaining in the soil, over 50% was within the surface layers of 20 to 30 cm depth, and reached barely detectable levels(less than or equal to 2.5 kg N ha(-1)) beyond 60 cm. Nitrate concentrations in soil-water samples collected at 90-cm did not show any significant increase with the split N applications in 1995/96, the wettest growing season, under commonly used rates of application, 160 kg N ha(-1). The results therefore suggests that, for rainfed wheat grown on heavy textured soils (Eutric Vertisol) of the Mediterranean Region, leaching losses of fertiliser N below 90 cm (plant rooting depth) during the wheat growing season when evapotranspiration accounts for most of the seasonal average rainfall (500 to 600 mm), were essentially nil. However, further work is needed to assess how the N retained in the soil after harvest of wheat will effect N utilisation of subsequent crops and likely losses to environment in future years.