This study was carried out in the summer seasons of 2010, 2011, 2012 and 2013, respectively, at Shandaweel Agriculture Research Station, Agri. Res. Center. The main objectives of the present investigation are to improve population Giza₁₀₂ of sunflower through development of synthetic varieties by S₁ per se family, Half-sib and Top-crosses recurrent selection procedure. In 2013 season, the first cycle of selection (C1) per se, H.S and T.C selection methods were evaluated against the original population Giza₁₀₂ and check variety Sakha53 at Shandaweel and Assiut Agric. Res. Sta. Randomized Complete Block Design (RCBD) with 3 replications were used in the two locations, to measure the actual gain from selection. Data were recorded on days to 50 % flowering, days to maturity, plant height, stalk diameter, head diameter, 100-achene weight, achene yield/plant, achene yield/plot and oil content. Highly significant differences between the two locations were detected for the studied traits. The combined means over two locations of oil content were 42.00, 43.37, 41.37, 39.91 and 39.98 (%), for S₁-cycle. H.S, T.C, original pop. and check variety, respectively. The differences between the three methods selection were significant, except between S₁ and T.C-cycle was not significant, but the differences between the first cycle selection, original pop. and check variety were significantly increased, except between the T.C-cycle and original pop. and check variety was not significantly increased. With respect to oil content, the observed gain from selection for S1 and H.S family selection methods (5.24% and 8.67%), respectively, were higher than those of T.C (3.66%) selection method. Finally, the S₁ per se and H.S selection methods were more effective in improvement the population Giza₁₀₂ of sunflower than the T.C method.