ESTIMATION OF HETEROSIS AND COMBINING ABILITY IN F1 AND F2 GENERATIONS FOR YIELD AND FIBER TRAITS IN COTTON

This study was done at Sids Agricultural Research Station, Beni-suef Governorate, Agric. Res. Center, Egypt, during three growing seasons (2018-2020). Six cotton varieties (Gossypium barbadense L.) comprised four Egyptian varieties (Giza85, Giza90, Giza95 and Giza91), Pima S6 (American Egyptian variety) and Karshenky (Russian variety) were crossed in half diallel manner to obtain 15 crosses. Parents and their 15 crosses in F1 and F2 generations were evaluated to define the best combiners to be used in breeding programs to improve cotton traits. Results showed highly significant genotypic differences for all the studied traits. General combining ability (GCA) and specific combining ability (SCA) were highly significant for most studied traits in F1 and F2 generations. Giza85 variety gave significant GCA effects for fiber traits, while, Pima S6 had significant GCA effects for yielding traits. Four crosses (Giza85 x Giza90, Giza85 x Karshenky, Giza95 x Karshenky and Giza95 x Pima S6) showed significant positive SCA effects for most yielding traits in F1 and F2 generations, while the cross Giza85 x Giza91 had significant positive SCA effects for some fiber traits. These genotypes were recommended for using in cotton breeding program to enhance the yield potential and fiber quality in the Egyptian cotton. Moreover, Giza85 x Giza90 and Giza95 x karshenky F1 crosses showed the best mid- and better-parent heterosis for yield traits, while Giza85 x Giza91 and Giza95 x karshenky crosses had the best heterosis for fiber quality traits. The non-additive genetic variance was larger than the additive variance in F1 and F2 for most of the studied traits except for number of bolls, seed and lint cotton yields/plant in both generations which were controlled by non-additive gene action (dominance or epistasis). Broad-sense heritability showed high values for all the studied traits, while narrow-sense heritability had low values for most of traits.