Induced polygenic variability in M2 generation and its relationship with production of high-yielding mutants in finger millet

Seeds of two finger millet varieties, VR 708 and GPU 26 were treated with three doses each of gamma rays (150, 300 and 450 Gy), ethyl methane sulphonate (0.15, 0.30 and 0.45%) and nitroso guanidine (0.015, 0.030 and 0.045%) in addition to two combination treatments of gamma rays 300 Gy + EMS 0.30% and gamma rays 300 Gy + NG 0.030%. The M1 generation was harvested as treatment bulk and the M2 to M4 generations were raised. In M2 generation, most treatment populations exhibited reduction in population mean and increase in population variance for all the six traits studied and the magnitude of such changes varied with mutagens, their doses and the variety. In general, greater shift in mean and variance was observed in treatments with higher doses of NG and EMS in case of VR 708 and higher doses of NG and combination treatments in case of GPU 26. Most mutagen treated populations showed wider range of variation than the parent variety and the variation was in both directions. Genetic advance estimates showed that selection in many M2 treatment populations would be effective in bringing about improvement in yield/plant and its direct components like tillers/plant, fingers/ear and finger length. Following selection among M2 plants and M3 progenies on the basis of higher yield, eight high yielding mutant cultures in VR 708 and nine mutants in GPU 26 were isolated in M4 generation. Vast majority of the high yielding M3 progenies and M4 mutant cultures were from the groups of M2 mutagenic treatments showing significantly higher population variance for yield/plant. Thus, selection of high-yielding M2 plants and M3 progenies in mutagenic treatments with much increased M2 variance for yield would be effective in isolation of high yielding micromutant cultures.