Sesame (Sesamum indicum L.), is believed to be first domesticated in India (Bedigian, 2004) and considered to be the oldest oilseed crop known to man (Duhoon et al, 2004). It is now grown in many parts of the world including tropical, subtropical and southern, temperate areas. Sesame seed is an important source of edible oil and also widely used as a spice. The seed contains 50-60% oil and 18-25% protein (Sabah El Khier et al., 2008). Sesame oil has excellent stability due to the presence of natural antioxidants such as sesamolin, sesamin and sesamol (Crews et al., 2006). Seeds with hulls are rich in calcium (1.3%) and provide a valuable source of minerals (Onsaard, 2012).

 

Total world output of sesame seed is 4.8 million tons, out of which 51.3% is contributed by Asia with the major share of India (Fao, 2013). Sesame is primarily a crop of developing countries growing by small farmers. Even though many high yielding varieties released in India, it has not contributed much to the oil scenario of the country. The progress in yield improvement is either stagnated or very slow during last 20 years because little attention is given on sesame development programmes in developing countries like India. The most powerful tool concerning to break this stagnated yield plateau in sesame is the development of hybrid varieties, but exploitation of hybrid vigour is a challenging concern of sesame breeder. Further, Novel combinations of beneficial alleles at multiple loci could lead to new potential for varietal improvement.

 

Earliness characters are of paramount importance in breeding for early maturing varieties/hybrids of oilseed crops in general and sesame in particular for better adaptation to climate change (Paroda, 2013). Early maturing varieties contribute significantly to increasing productivity (Saravanan et al., 2000a). Selection for maturation period can be effective using flowering period for improving uniform ripening capsule (Jamie et al., 2002). A suitable breeding methodology and the identification of superior parents are the most important pre-requisites for the development of early maturing and high yielding genotypes. For genetic improvement of the crop, the breeding method to be adopted is depending upon the nature of gene action involved in the expression of different characters. The genetic makeup of genotypes for quantitatively inherited traits can be well understood by the study of genetic parameters. Combining ability analysis has been utilized to know the nature and extent of gene action controlling expression of different characters including seed yield and would help in proper planning of a successful breeding programme. Several workers used combining ability analysis in sesame and identified superior combining parents (Saravanan et al., 2000a; Banerjee and Kole, 2009; Azeez and Morakinyo, 2014). Thus, among the biometrical tools, diallel analysis is widely used breeding procedure and more suitable to choose appropriate parents and crosses, and to determine combining abilities of parents in the early generation. Again, genetic diversity or allelic divergence among the parents is very important in selecting parents for hybridization programme to identifying heterotic crosses and obtaining desirable recombinants in the segregating generations.

 

The objectives of the present study were to assess the nature and magnitude of gene action controlling the inheritance of earliness characters and seed yield in F₁ generation, to identify good combining parents to be used as donors for improvement of traits and to identify crosses for development of early maturing hybrids and/or selection of superior segregates in advance generations of segregation.

 

1 Results

Analysis of variance for combining ability (Table 1) revealed that mean squares due to general combining ability (GCA) and specific combining ability (SCA) were significant for earliness traits and seed yield per plant in the F₁. The mean square for the GCA was higher than the mean square due to SCA for the all the traits. This revealed importance of both additive and non-additive gene actions in the characters. However, component of genetic variance indicated that estimates of σ²_gca were lower than those of σ²_sca and the predictability ratio was also below unity. Thus, nonadditive gene action was preponderant in for all the characters studied.

 

 

The gcaeffect was maximum and significantly negative for AT 158 followed by AT 177 in days to flowering initiation and days to 50 per cent flowering but in days to maturity it was observed in order of AT 177, AT 158 and RT 54 (Table 2). Apart from this, the highest and significantly positive gcaeffect was evidenced from the AT 177 followed by ABT 33, AT 192 and G.Til 2 for plant height, while in seed yield per plant, the RT 54 followed by AT 158 exhibited a significantly positive gca effect. The parent ABT 33 only had a highly significant gca effect for the reproductive period and plant height. The per se performance for studied characters distinctly characterized parents with differential values for different traits.

 

 

The sca estimates represent dominance and epistasis. Among 28 cross combinations evaluated, the highest sca effect in desirable direction was expressed by AT177×G Til 1 for days to flowering initiation and reproductive period, AT 177×RT 54 for days to 50% flowering, G Til 1×G Til 2 for days to maturity, AT 177×TKG 22 for plant height and AT 192× G Til 1 for seed yield per plant (Table 3). None of the crosses found good specific combiner for all the characters under consideration. However, the AT 177× G Til 1 and AT 192×G Til 2 for four characters and AT 177×ABT 33 and AT 192 × G Til 1 for three characters including seed yield per plant manifested significant specific combining ability values. The cross AT 177×RT 54 and AT 192×G Til 2 were good specific combiners for earliness with increased plant height. While the cross AT 192×G Til 1 was good specific combiners for seed yield per plant, early flowering and maturity.The top ranking five crosses based on per se performance and sca effect are listed in Table 4. Each parents of these crosses based on their gca effects were classified as high (H), average (A) and low (L) general combiners. Parents with significant desirable gca effects were considered as high combiners, parents showing non-significant estimates either with ‘+ve’ or ‘-ve’ sign were classified as average combiners and parents had significant gca effects in undesirable direction were classified as low combiners. The significant +ve sign for seed yield per plant, plant height and reproductive period and significant –ve sign for all the earliness traits were considered as in desire direction. Thus, a perusal of the table indicated that out of five best per se performing crosses, three crosses showed high sca effect for days to flowering initiation, days to maturity plant height and seed yield per plant and two crosses for days to 50% flowering and reproductive period.

 

There was no relationship between per se performance and sca effect in most of the crosses.

 

 

 

2 Discussion

The choice of breeding method primarily depends upon the nature and magnitude of gene action. The knowledge on combining ability and type of gene action for different characters helps to pinpoint the best combiners and proper planning of a systematic breeding programme. Again, genetic diversity or allelic divergence among the parents is very important in selecting parents for hybridization programme for identifying heterotic crosses and obtaining desirable recombinants in the segregating generations. Generally, GCA is associated with additive genetic effects, while SCA is attributed to the non-additive (dominance and epistasis) genetic effect (Banerjee and Kole, 2009). Additive (intra-allelic) and additive× additive (inter-allelic) types of gene effects are the fixable portion of the total genetic effect (Saravanan et al., 2000a) and are, therefore, useful for developing homozygous lines in self pollinated crops. In the present study, predictable ratio revealed that all the characters studied were predominantly controlled by non-additive gene effects (Table 2). However, additive portion was also important as indicated by significant test of GCA variance for all the traits. Under such circumstances, breeding procedure which can increase genetic recombination and also help accumulation of favourable genes in later generation may prove most effective in improving seed yield and earliness. Importance of additive and non-additive (Saravanan et al., 2000a) as well as additive and dominance (Saravanan et al., 2000b) gene actions were reported for earliness in sesame.  Seed yield and plant height also reported to be governed by both additive and non-additive gene actions (Singh, 2007; Toprope, 2008; Sharmila and Ganesh, 2008) while Yamanura et al. (2009) reported predominance of non-additive gene action for  days to 50 per cent flowering, days to maturing, plant height and seed yield per plant in sesame. The most appropriate method for exploitation of preponderance of non-additive gene effects would be recurrent selection (Saravanan et al., 2000a) as it concentrates favourable genes together in a single genotype through intermating and selection.

 

It was observed that the parents with high gca effects not always exhibited high mean values and vice versa. This suggests that considering per se performance and gca effects together would be reliable practice while selecting parents in breeding programmes. In this context, parent AT 158, AT 177, RT 54 and ABT 33 may be judged as good parents. The multiple crosses involving these parents would be an appropriate practice to be employed in the development of hybrids and or selection of superior recombinants in the segregating generation to isolate early maturing genotypes with higher yield.

 

The gca effects of the parents involved in the crosses showing high per se and sca effect (Table 4) revealed that majority of the crosses involved either H×A or H×L gca parents and a few H×H and L×L combinations. Manifestation of high sca effects in the cross involving H×H gca parents indicates sizeable additive×additive gene interaction. The H×L cross may be due to additive×dominance while L×L cross might be ascribed to dominance×dominance type of non-allelic interaction. Thus, it is clear that the superior performance of most crosses in the present study was largely due to epistatic interaction (Saravanan, et al., 2000a).Nevertheless, out of five high sca crosses, four crosses were of H×H and one cross was of H×L gca parents for plant height. This indicated two types of epistatis-additive×additive and additive×dominance which are largely fixable. In such situation, simple selection or modified pedigree selection would be fruitful. However, as importance of both additive and non-additive gene action revealed in the present study, delay selection in later generations by adopting intermating of segregants followed by recurrent selection would be a rewarding approach (Saravanan, et al., 2000a) for improvement of seed yield and earliness characters in sesame.

 

Superior cross combinations for seed yield per plant, namely, AT 192×G Til 1 exhibited significantly desirable sca effect also for days to 50% flowering and maturity which involved either A×A or A×L gca parent for these traits. Such a relationship between gca and sca effects indicates the importance of dominance×dominance type of epistatic interaction and also reveals potentiality of low gca parents to express high sca effects in cross combination. Therefore, it may not always be necessary to attempt crosses between high×high gca parents, but low gca parents can also manifest high scaeffects, which is attributable to non-fixable interaction effects and could be exploited through heterosis breeding.

 

Critical examination of Table 2&3 indicated that AT 158×AT 177 for days to flowering initiation as well as AT 158×RT 54 for days to maturity and seed yield per plant expressed non-significant sca effects though both the parents having significant gcaeffects. This indicates the presence of additive gene systems in both patents for these characters. Recombination breeding can exploit additive gene action which is fixable. The segregation of these hybrids is likely to throw more recombinants possessing favourable additive genes from both the parents (Sakila et al., 2000).

 

The present study has identified desirable parent viz., AT 158, AT 177, RT 54 and ABT 33 as good general combiners for different characters related to earliness and seed yield. The importance of additive but mainly non-additive gene effect is highlighted. Improvement in earliness combined with seed yield should be possible by adopting intermating of segregants followed by recurrent selection. Considering sca effects, the cross AT 192×G Til 1appears to be promising for earliness and seed yield. Desirable specific combiner cross AT 177×ABT 33 involving parents with gca A×A for seed yield, H×A for days to maturity  and H×H for plant height could be utilized for selection of superior genotypes in advance segregating generations. To exploit additive effects of the cross AT 158×AT 177 for days to flowering initiation, and AT 158×RT 54 for days to maturity and seed yield per plant whose sca effect was non-significant but their parents were high general combiner, recombination breeding should be a rewarding approach.

 

3 Materials and Methods

3.1 Plant material used

Eight sesame genotypes were crossed in a half diallel mating design (excluding reciprocals) during summer 2011 to produce F₁ crosses. The parent genotypes were G Til 1 and G Til 2, both white seeded and most widely adopted varieties from Gujarat (Western India); RT 54, early maturing brown seeded commercial cultivar from Rajasthan (Western India); TKG 22, commercial cultivar from Madhya Pradesh (central India) is using as national and zonal checks in different breeding programmes in India, rest of four were promising lines namely AT 158, AT 177, AT 192 and ABT 33 all from Gujarat (Western India). In addition to differences in maturity time, importance was given to choose the parents representing distinguishable characteristic features. AT 158 was bolder in seed size, AT 177 possesses longer capsule, higher number of seeds per capsule and high oil content. AT 192 was also a longer capsulated early line, while ABT 33 was of black seeded tall plant. TKG 22 was small seeded old cultivar, whereas RT 54 was the cultivar of short plant stature with short capsule length and low oil content. Mean performance of the eight parental genotypes for different characters is presented in Table 1.

 

3.2 Experimental details

The experiment was conducted in rainy season of 2011 (July to September) at the Agricultural Research Station, Junagadh Agricultural University, Amreli located at latitude of 21.36^oN, longitude of 71.13^oE and an altitude of 130 m above mean sea level representing semi-arid area of Gujarat, India. The soil type of experimental site was medium black, low in organic carbon, nitrogen and phosphorus and rich in potash with pH of 7.0 and EC of 0.23 dSm^-1.The site received annual rainfall of 804 mm in 38 rainy days during the period of crop growth with a respective minimum and maximum temperature of 23.0 and 32.5°C, respectively. The average relative humidity during crop growth period was 76.5%. The 28 F₁ and their eight parents harvested from summer 2011 crossing programme were evaluated in randomized block design with three replications during rainy season of same year. Each entry was hand sown in single row of 3.5 meters following row to row spacing of 60 cm. Plant to plant distance of 10 cm was maintained by hand thinning at 20 days after sowing. The soil of experimental area was fertilized at the rate of N 25 kg·ha^-1 and P 25 kg·ha^-1 as basal dose before sowing. Additional N 25 kg·ha^-1 was applied as topdressing at 30~35 days after sowing. Normal recommended cultural practices were followed during experimentation to raise a successful crop for better phenotypic expression of characters.

 

3.3. Data collections

Randomly selected five plants from each plot were used for recording plant height (cm), and seed yield per plant (g). While days to flower initiation, days to 50% flowering, days to maturity and reproductive period were determined on plot basis. Plant height (cm) was measured in cm from ground level to the tip of main shoot at maturity. Dry and clean seed in gram harvested from selected plant of each genotype was considered as seed yield per plant (g). Flowering initiation was the days from sowing to the date of first flowering opening in the plot. Date on which flowering opening observed on 50% plants in a plot was used to compute days to 50% flowering. Days to maturity were the duration from date of sowing to the date on which approximately 90% of the green capsule in a plot turned yellow. Reproductive period was recorded as days taken from date of 50% flowering to date of maturity (Olowe, 2007).

 

3.4 Statistical analyses

Data were analyzed to estimate combining ability following method II (parents plus one set of crosses), model I (fixed effect model) as suggested by Griffing (1956) using SAS statistical software (SAS, 2004). The statistical model for the mean value of a cross (i×j) is:

Y_ij = μ + g_i + g_j +S_ij + l/b ∑∑e_ijkl

Where, Yij = mean of (i×j) th cross over replications k (k = 1, 2,…,r)

μ = general mean; gi and gj = GCA effects of i^th and j^th parent, respectively

sij = SCA effect for the cross involving i^th and j^th parent

l/b ∑∑e_ijkl = Mean error effect

 

The estimates of genetic components were obtained based on the expectations of the mean squares as:

σ²_gca = (MSGCA – MSerror)/2r

σ²_sca = (MSSCA – MSerror)/r.

Where, MSGCA = mean square due to GCA

MSSCA = mean square due to SCA

MS_error = error mean square

r = number of replications.

 

The relative importance of variances due to GCA and SCA was compared following predictability ratio (Baker, 1978).

Predictability ratio =2 σ²_gca /2σ²_gca+ σ²_sca

 

Authors'contributions

AK carried out genetic studies and performed the statistical analysis. BA conceived of the study, participated in its design and drafts the manuscript. All authors read and approved the final manuscript.

 

References