Combining ability analysis for yield and its component traits was undertaken with five lines and 30 testers using Line x Tester mating design. The estimated components of variance for Specific Combining Ability (SCA) effectswere larger in magnitude than the General Combining Ability (GCA) effects for all the characters, indicating the predominance of non-additive gene action for control of these characters. The predominance of variance due to SCA over GCA showed that non-additive type of gene action was involved for all the traits studied. The estimates of GCA effects indicated that the parent ICMA 94111A1 was a good combiner for all the traits studied except the grain yield per plant. An overview of the mean performance and the GCA effects of parents revealed that the ICMA 88004A and testers PT 5447, PT 5451 and PT 5480 were identified as good parents for further breeding program. The ICMA 88004A x PT 5433, 732A x PT 5172, ICMA 88004A x PT 5480, 81A x PT 5656 and PT 5054A x PT 5451 were identified as superior hybrids and can be effectively utilized for hybrid breeding programs.

Taxonomically, Pennisetum glaucum (pearl millet) belongs to the family Poaceae, sub-family Panicoideae, section Penicillarium and possesses 2n = 2x =14 chromosomes. India is the largest producer of pearl millet (Pennisetum glaucum (L.) R. Br.), both in terms of area (9.1 m ha) and production (7.3 mt), with an average productivity of 780 kg/ha during the last five years. Nearly 50% of the millet area is under hybrid cultivars. Though the bajra hybrids are being developed by utilizing different cytoplasmic sources, the practical utility of these newly developed cytoplasmic male sterile sources depend on the combining ability and vigor expressed by the hybrids. Hence, the basic objective for any breeding program is to select the parents with good combining ability to produce superior hybrids. This study aims at estimating the combining ability and understanding the nature of gene action for a set of male sterile and inbred lines.

The analysis of variance for combining ability, using line x tester mating design for all the eight characters, is furnished in Table 1 and combining ability effects for grain yield and its components of pearl millet are furnished in Table 2. The estimates indicated that the lines, testers and line x tester interactions were significant for all the characters studied. The estimated components of variance for SCA were larger in magnitude than GCA for all the characters, indicating the predominance of a non-additive gene action to control these characters Navale et al. (1991), and Dhuppe et al. (2006).

Combining Ability Analysis, Pearl Millet, Pennisetum glaucum,Cytoplasmic, Panicoideae, Penicillarium, General Combining Ability, GCA, Specific Combining Ability, SCA, Kempthorne, Genetic Statistics, Randomized Block Design, Plant Breeding and Genetics.