| Gene effects The results of A, B and C scaling tests are given in Table (3). For all traits except number and weight of kernels per spike there were highly significant values for one or more of these tests. Hence, the additive-dominance model was inadequate indicating that epistasis was involved in the inheritance of most traits studied. The joint scaling test is more powerful than any of the other test in detecting epistasis since it uses information from all six populations. A three-parameter model (m), (a) and (d) proved to be satisfactory in explaining the genetic differenes. The estimates of the three-parameters and the test of goodness of fit for the model are given in Table (4). Chi-square (x2) values for the differences between observed and expected population means provided significant values. These results were agreed with A, B and C scaling tests in showing that the additive-dominance model was inadequate. ABO-ELENIN and GOMMA (1972) studied the nature of gene effects for plant height by using scaling tests and they detected inadequace of additive-dominance model on this trait. The estimates of dominance genetic effects (d) were larger than the additive genetic effects (a) for all traits except date headed and kernels weight per spike and always negative. The signs of (a) and (ad) depends upon which parent is considered as P1 or P2. The highest dominance effects (d) were involved for plant height and number of kernels per spike. The additive genetic effects (a) were smaller in magnitude but significant for all traits exept number of kernels per spike. It can be concluded that both additive and dominance effects were involved in the behavior of all traits except number of kernels per spike which controlled by dominance effects. Because of epistasis, the three-parameter model was not sufficient to explain the genetic variation. Therefore the six-parameter model (Table 4) was invoked to determine the type and magnitude of gene action. The estimates of dominance effects (d) from the three-parameter model were smaller than the estimates from the six-parameter model. The six-parameter analysis showed that the magnitude of dominance effects (d) were greater than the estimates of additive effects (a). While the dominance x dominance effects (dd) were relatively larger in magnitude and more important than the other two types i.e. (aa) and (ad) for all traits except plant height. It can be concluded that dominance x dominance epistatic effects (dd) as well as dominance (d) played the major contributions to the inheritance of most traits studied. These results agreed with that obtained for kernel weight (SUN et al., 1972), grain yield (AMAYA et al., 1972) and for harvest index (ALI and EL-HADDAD, 1978). On the other hand, additive (a) and additive x dominance epistasis (ad) played an important part in the inheritance of some traits such as plant height and number of spikelets per spike. Similar results were obtained by BHATT (1972) and KETATA et al. (1976). The estimates of epistasis, dominance and additive gene action may have been influenced by genotype-environment interactions in both three and six-parameter model. |
| <-- Back | | | --> Next |