To study the phenotypic stability of barley, sixteen different barley genotypes were grown under 9 different  environments [three different levels of nitrogen (25, 40, and 55 kg N/faddan) X three successive seasons] in the Agricultural Research Station of  Faculty of Agric., CairoUniversity, Giza, Egypt. A lattice square design with five replicates was used for each yield trial. The stability parameters: regression coefficients (b_i), mean squares due to deviations from regression (S²_d), ecovalence (W_i), and the two non-parametric measure S¹ and S³ were used. The Egyptian genotype (Giza 123) showed the highest mean grain yield (0.49 Kg  m^-2) and the highest values of b_i for grain yield / m² (b_i= 2.02) and for number of grains / spike (b_i= 4.45). These results showed that the high yielding genotypes exhibited high response and more adaptability to good environmental conditions. Incontrast, the genotype No.8 exhibited the lowest yielding ability (0.36 Kg m^-2) among all studied genotypes. Highly significant positive correlation coefficients were detected between number of grains / spike and each of S²_d, b_i, W_i, S¹ and S³. Incontrast the correlation coefficients between the means of the other three studied traits (grain yield / m², seed index and spike length) and stability parameters  were not significant. The observed correlation among stability parameters indicated  that the ecovalence is highly significant associated with mean squares due to deviations from regression (S²_d) and two non-parametric measures (S¹ and S³). Highly significant correlation coefficient was detected between S¹ and S³ for all studied traits. Our results showed that the production response index (b_i) and other stability statistics could be used in addition to mean yield by breeder in barley selection programmes when G x E is present.