15N-labelled fertilizer was applied at different rates (0, 30, 60, 90 kg N ha−1) and in different forms (urea or ammonium sulphate) to wheat grown in Syria in three seasons (1991/92, 1992/93 and 1994/95).

Recovery of 15N-labelled fertilizer in the above-ground crop at harvest was low (8–22%), with the amount of 15N-labelled fertilizer recovered in the crop increasing as the rate of application increased. Fertilizer application caused a significant increase in the amount of unlabelled soil N in the crop, suggesting that the application of N fertilizer caused a ‘real’ added nitrogen interaction. Recovery of 15N-labelled fertilizer in the crop was unaffected by the form of the fertilizer.

On average 31% (14–54%) of the 15N-labelled fertilizer remained in the soil at harvest, mostly in the 0–20 cm layer. At the lowest application rate (30 kg N ha−1) most of the residual fertilizer was as organic N, but at the higher application rates (60 and 90 kg N ha−1), a greater proportion of the 15N-labelled fertilizer was recovered as inorganic N, presumably as the result of top-dressing N in dry conditions in the spring. The amount of 15N-labelled fertilizer remaining in the soil increased as the fertilizer rate increased, but was unaffected by the form of fertilizer applied.

Losses of 15N-labelled fertilizer were large (>35%), probably caused by gaseous losses, either through volatilization of N from the calcareous soil, or through denitrification from wet soils rich in organic residues.

N fertilization strategies in the West Asia/North Africa (WANA) region should take note of the low recovery of N fertilizer by the crop in the season of application, and the resultant large quantities of residual fertilizer.