Sunflower is an important edible oil crop. Its yield on dry lands of West Asia-North Africa is low due to limited rainfall at the time when this crop is in full growth during May-June, thus supplementary irrigation (SI) must be used. However, this region lacks river/canal water and irrigation, when possible, is from underground water storages, which, due to their quantitative scarcity, have to be used judiciously. This trial was conducted to assess the minimum SI application needed to produce optimum economic yields. The study was made for three years (1995-1997) at Tel Hadya, ICARDA, in northern Syria, on a soil characterized as fine clay (montmorillonitic, thermic Calcixerollic Xero chrept) with a pH around 8.0. We tested two open pollinated varieties, HO-1 and Record, and three water level treatments, rainfed, 50% of crop water requirement and 100% of crop water requirement. Effects of environment (years), variety and water level were observed on plant height, head and stem diameter, 200-seed weight, leaf area and sunflower grain yield. The results show that plant height, head diameter, stem diameter and leaf area differed significantly due to year effects while seed weight and yield differed non-signifi cantly. The cultivars were significantly different only regarding seed weight and stem diameter. Grain yield obtained under rainfed conditions was highly significantly lower than those obtained under SI, with both lower and higher quantity of irrigation water. The average yield increase with 50% SI was 150% and with 100% SI was 312%. However, the yield difference between the two irrigation levels was 66%. This trial showed that 50% SI irrigation level can increase the yield substantially and economically, and can be resorted to in the absence of water for full irrigation.