(go to KOMUGI Home) (go to WIS List) (go to NO.80 Contents)


According to the data presented in Table 2, low yielding parent Buc's and Blue Silver combine well to exhibit better parent heterosis, which indicated the preponderance of additive gene action. The hybrids P1x P2 and P4 x P1 are likely to produce high yielding progenies in early generations due to better specific combining ability (Table 4), this indicates that yield per plant is the expression of additive and dominant genes. This is confirmed by the findings of Lupton (1961), who found that certain crosses with large standard deviation but low mean yield displayed greater promise than those with high yield. According to East (1936), hybrid vigour may also be due to accumulation and fixation of favourable genes, the maximum number of which is brought together in the F1 hybrids, but the intensity of action of certain genes which manifest heterosis may be very low as a result of inbreeding.

Heterosis for these yield components has an important relationship with heterosis for grain yield. The crosses expressing significant and positive heterosis for yield per plant had significant and positive heterosis for some yield components. In F1 generation significant and positive heterosis, particularly for spike length, spikelets per spike, seeds per spike, yield per spike and seed index was most frequently associated with significant and positive heterosis for yield per plant (
Table 2). Similar positive relationship between heterosis for yield per plant and heterosis for yield components was reported by Larik et al. (1988, 1992).

When the heterosis for the crosses was compared with their SCA effects, it was observed that both were positively related. The crosses P1
x P2, P2 x P4, P4 x P6, P5 x P6, P2 x P1, P3 x P1, P4 x P1 and P5 x P1 had significant estimates of both SCA effects and heterosis for yield per plant (Table 4). Significant estimates of both heterosis and SCA effects suggest predominance of non-additive gene action for yield per plant in these crosses. Selection through conventional breeding methods would not be effective in these crosses, alternatively development of hybrid variety might be a good choice.

2. Combining ability

The analysis of variance for general combining ability (GCA), specific combining ability (SCA), and reciprocal effects (RE) are presented in
Table 5. Both GCA and SCA variances were highly (P<0.01) significant for plant height, spike length, seeds per spike and single plant yield, whereas GCA was only significant (P<0.05) for fertile tillers per plant. RE were highly significant (P<0.0 1) for plant height, seeds per spike and single plant yield. GCA variance contains additive and additive x additive epistasis while SCA variance contains dominance and additive x dominance, dominance x dominance epistasis (Griffing 1956; Baker 1978), so the significant estimates of GCA and SCA variances suggest that both additive and non-additive gene actions were involved in controlling these characters in the present materials. The variance for GCA were larger than those of SCA for all the traits, which suggest that the major portion of genetic variability in the base population was additive in nature. Higher estimates of non-additive genetic variance were noticed only for seeds per spike (Table 5). These results suggest that the yield components were predominantly controlled by additive gene action. But seeds per spike were mainly controlled by non-additive gene action.

<--Back | -->Next

(go to KOMUGI Home) (go to WIS List) (go to NO.80 Contents)