Generation mean analysis for grain yield and its contributing traits in pearl millet [Pennisetum glaucum (L.)]

The experiment, consisting families (P1, P2, F1, F2, BC1 and BC2) of 3 crosses of pearl millet viz., R-16419 × R-15114, R-15134 × R-15762, and R-15762 × ASRT-111. Joint scaling test suggested that nodes per plant and days to physiological maturity were adequate for 3 parameter model. Six parameter models revealed that both additives (d) as well as additive × additive (i) type of gene effects were significant for seed setting in cross ‘R-15134 × R-15762’. Duplicate type of epistasis was found for plant height, grain yield per plant in all three crosses of pearl millet, flag leaf length, 1000 seed weight in cross ‘R- 16419 × R-15114’, tillers per plant, panicle length, seed setting, middle leaf temperature, harvest index in cross R-16419 × R-15114 and tillers per plant, panicle length, flag leaf length, middle leaf temperature, 1000 seed weight in cross of R-15762 × ASRT-111. Significant role of epistasis revealed that both additive as well non-additive effects involved in the expression of yield and its attributes, thus, reciprocal recurrent selection would be an ideal method for population improvement. Situation like duplicate type of gene action, breeding procedures involving multiple crosses, bi-parental crosses may be restored to get transgressive segregates.