Degree of Dominance and Generation Mean Variance Analysis for Biometrical Traits in Sesame ( Sesamum indicum L.)

The experiment was conducted with six generations derived from five inter varietal crosses TMV 5 x KS99037(C1), TMV 5 x KS990812 (C2), TMV 5 x KS990813 (C3), TMV 5 x KS99153 (C4) and TMV 5 x Cordebergea (C5) involving six genotypes of sesame to study the components of variances and degree of dominance for different traits. The observations on days to first flowering, days to maturity, plant height, number of primary branches, number of capsules per plant, capsule length, number of seeds per capsule, 100 seed weight and seed yield per plant were recorded for the selected individual plants. The variances estimated from different generations were partitioned into different genetic components of variation. The degree of dominance was greater than unity in TMV 5 x KS99037 for days to maturity, in TMV 5 x KS990812 for days to first flowering, in TMV 5 x KS990813 for days to first flowering and days to maturity, in TMV 5 x Cordebergea for number of seeds per capsule which showed over dominance for the respective traits. Hence, it is suggested for adoption of heterosis breeding for these crosses in sesame. The cross TMV 5 x KS99037 for days to first flowering, number of branches per plant and oil content and TMV 5 x KS990813 for plant height and number of capsules per plant showed partial dominance for the respective traits. The recombination breeding followed by selection in the later generation will improve the respective traits. The traits, plant height, number of capsules per plant and 100 seed weight were largely controlled by both additive and dominance variance. The pedigree or recombination breeding followed by selection in later generations would improve these characters. The environmental variance for the characters days to maturity, number of branches per plant, number of capsules per plant, capsule length, number of seeds per capsule and 100 seed weight revealed that environment plays an important role for the expression of these traits.

Sesame (Sesamum indicum L.) is an important edible oil seed crop of India which occupies the third rank next to groundnut and mustard. By virtue of excellent quality of its oil it is referred to as the "Queen of the oil seed crops". Sesame seed is of high economic importance due to its oil content, nutrient content and protein. Sesame oil is of immense importance in Aurvedic Medicines, daily household purposes and in religious ceremonies. But the productivity of this crop in India is very low when compared with the world average. For breaking the present yield barrier and evolving varieties with high yield potential, it is desirable to combine the genes from genetically diverse parents. The success in identifying such parents mainly depends on the gene action that controls the trait under improvement. Formulation of a comprehensive breeding programme for the improvement of any crop largely depends upon the nature of gene action involved for any particular trait to be improved. The development in statistical genetics have made possible to study the various facets of the operation of quantitative genes and to use this information in formulating appropriate breeding strategy to effect genetic improvement of a particular trait. The present study therefore was aimed at studying the genetics of important quantitative characters including seed yield, so as to formulate suitable breeding strategy.

Materials and Methods
The experiment was conducted with six generations viz., P1, P2, F1, F2, B1 and B2 derived from five inter varietal crosses TMV 5 x KS99037(C1) TMV 5 x KS990812 (C2), TMV 5 x KS990813 (C3), TMV 5 x KS99153 (C4) and TMV 5 x Cordebergea (C5). These five crosses were effected in a crossing block to obtain the hybrid seeds. The F1 plants were back crossed with the parents viz., P1 and P2, keeping the F1 as the ovule parents to get their respective back cross families B1 and B2. The F1s were also grown separately to obtain F2 seeds. With the reserved seeds of direct F1 hybrids of the five cross combinations, seeds for all the six generations viz., P1, P2, F1, F2, B1 and B2 were obtained. All the six generations were raised at Department of Oilseeds, Tamil Nadu Agricultural University, Coimbatore during rabi, 2005. The parents and F1s were planted in four rows of 6 m length. The B1 and B2 were planted in 6 rows of 6m length, while the F2 of each crosses were raised in 12 rows of 6m length. The inter and intra row spacing was maintained at 30 x 30 cm. Fifty plants each from parents and F1 s , 100 plants from B1 and B2 generations and 275 plants from F2 populations were randomly selected.
Observations on days to first flowering, days to maturity, plant height, number of primary branches, number of capsules per plant, capsule length, number of seeds per capsule, 100 seed weight and seed yield per plant were recorded for all these selected individual plants. The variances estimated from different generations were partitioned into different genetic components of variation viz., nonheritable variance due to environment (E), fixable variance due to additive genes (D), non fixable variance due to dominance (H), co variance of additive and dominance effects (F) and dominance ratio followed by Mather and Jinks (1977).

Results and Discussion
The estimates of different genetic components of variances and their ratios for yield and its components were presented in Table 1.
Days to first flowering: The variance due to environment was lower than the additive and dominance variances in only two crosses (C3 and C5), but was higher than both the variances in the cross C1. The additive variance was negative in three crosses and it was positive in two crosses (C1 and C3). The dominance variance was positive in all the crosses except C2 and it was higher than the additive variance in all the crosses except C1. The covariance of additive x dominance (F) component was positive in three crosses C1, C4 and C5. It indicated the predominance of dominant alleles in the parents of these crosses for this trait. The dominance ratio was recorded as more than unity in the cross C2 and C3 which indicated the over dominance gene action which inturn showed that the heterosis could be exploited from these crosses for days to first flowering. The cross C1 showed partial dominance gene action since it recorded the dominance ratio of less than unity. Dominance gene action was reported by Godawat and Gupta (1995) and Ramesh et al. (2000) for days to flowering.
Days to maturity: The environmental variance was lower than the additive and dominance variances in two crosses namely, C3 and C5. But the cross C2 was highly influenced by the environment. The additive variance was positive in two crosses (C1 and C3) and it was negative in three crosses where as dominance variance was positive in all the five crosses. The F component was positive in 75 C1, C4 and C5 but it was negative in C2 and C3 crosses. The dominance gene action was greater than the additive gene action in all the five crosses for days to maturity. The dominance ratio was recorded as more than unity in the crosses C1 and C3 which indicated the over dominance gene action. These crosses may be utilized to exploit heterosis. Similar result was reported by Sivagamy (2003) for this trait.

Plant height:
The environmental effect on this character was low compared to additive and dominance components in the crosses C1, C2 and C5. Additive variance was positive in all the five crosses and it was higher than dominance variance in three crosses viz., C1, C2 and C3. The dominance variance was negative in all the crosses except C3. The covariance of additive x dominance was positive in all the crosses except C2. It can be assumed that the parents of these crosses carried more dominant alleles, but the dominance ratio was less than one for the cross C3 which indicated partial dominance. It may be suggested that the character plant height can be improved through pedigree breeding followed by selection.

Number of branches per plant:
The variance due to environment was lower than the additive and dominance variances in two crosses viz., C1 and C2 and it was high in the remaining three crosses. The additive variance was positive in all the crosses while the dominance variance was negative in all the crosses except C1. The F component was positive in C1 and C3, but it was negative in the remaining crosses. The dominance variance was higher than the additive variance in all crosses except C1. The dominance ratio was recorded as less than one for the cross C1 indicating the partial dominance gene action for this trait. The heterosis can be exploited from the cross C1 since the parents of this cross also carried more dominant alleles for this trait. The same result was already reported by Vidhyavathi (2002).

Number of capsules per plant:
The cross C5 was less influenced by the environment compared to all the other four crosses for this trait. Additive variance was positive in all the crosses except C1, whereas the dominance variance was positive in only two crosses C1 and C3. The dominance ratio was less than one in the cross C3 which indicated the partial dominance gene action for this character. The F component was positive in all the crosses except C2. In general, for all the crosses, the additive variance was higher than the dominance variance and hence it may be largely controlled by additive gene action. This indicated that the individual genotype could be evaluated readily from its phenotypic expression. Simple selection or simple recurrent selection would be more effective for this trait. The additive gene action for this trait was also reported by Das and Gupta (1999) and Ramesh et al. (2000). Capsule length: This character was less influenced by the environment. The additive variance was positive in C3 and C5 while the dominance variance was positive in C1, C2 and C4. The covariance of additive x dominance was positive in C2, C4 and C5 but it was negative in the remaining crosses. The dominance variance was higher than the additive variance in four crosses; hence recombination breeding followed by selection in the later generation will improve this trait. Senthil kumar and Ganesan (2002) reported dominance gene action for this trait.

Number of seeds per capsule:
The crosses C1 and C5 were highly influenced by the environment. The additive variance was negative in two crosses viz., C1 and C2 and it was positive in the remaining crosses while, dominance variance was positive for C1 , C2 and C5 and negative for the remaining crosses. The covariance of additive x dominance was positive in C2, C3 and C5. It was assumed that the parents of these crosses carried more of dominant alleles. The dominance of the cross C5 was also confirmed by the dominance ratio which was very high, more than unity in the cross C5, indicating the over dominance gene action and exploitation of heterosis for this character through this cross could be possible. The non additive gene action for this trait was earlier reported by Solanki and Gupta (2001) and Deepa Sankar and Ananda Kumar (2003). 100 seed weigh: The environment influence was observed for this trait.The additive variance was positive in C2 and C5 but the dominance variance was positive in C1, C3, and C4. The F component was negative in all the crosses except C1. There was not much difference between dominance and additive variances. Pedigree breeding followed by selection would improve this character. Additive variance for this trait was reported by Ramesh et al, (2000) and Sivagamy (2003).
Seed yield per plant: Seed yield was not much affected by the environment for these crosses and the environment variance was lower than the additive and dominance variances. The additive variance was positive for C2, C3 and C5 while the dominance variance was positive for C1 and C4. The F component was positive for C1, C3 and C4 and it was negative for the remaining crosses. Dominance variance was higher than the additive variance for all the crosses. Recombination breeding with selection in later generation might be followed for the improvement of this trait. Dominance gene action for seed yield was reported by Saravanan and Nadarajan (2003) and Senthil Kumar et al. (2004) Oil content: The dominance variance was slightly higher than additive variance and it was positive for C4 only while the additive variance was positive in the cross C2, C3 and C5 and it was negative in C1 Environmental influence was comparatively low for this trait in three crosses namely C2, C3 and C5. The F component was negative for all the crosses indicating the presence of recessive alleles in the parents. The dominance variance for the cross C1 was less than one which showed the partial dominance gene action. Recombination breeding with selection in later generation might be followed for the improvement of this trait. Additive variance for this trait was reported by Devasena et al. (2001) and Sivagamy (2003).
In the present study, Environmental variance for days to maturity, number of branches per plant, number of capsules per plant, capsule length, number of seeds per capsule and 100 seed weight showed that environment plays an vital role in the expression of these traits. Plant height, number of capsules per plant and 100 seed weight were controlled by both additive and dominance variance. Pedigree or recombination breeding followed by selection in later generations would improve these characters. The cross TMV 5 x KS99037 for days to first flowering, number of branches per plant and oil content, TMV 5 x KS990813 for plant height and number of capsules per plant, showed partial dominance for the respective traits. Recombination breeding followed by selection in the later generation will improve the respective traits. Degree of dominance was greater than unity in TMV 5 x KS99037 for days to maturity, TMV 5 x KS990812 for days to first flowering, TMV 5 x KS990813 for days to first flowering and days to maturity, TMV 5 x Cordebergea for number of seeds per capsule, showed over dominance for the respective traits. Hence, it is suggested for adoption of heterosis breeding for these crosses in sesame.