Abstract: Combining ability analysis of eight Brassica napus L. genotypes was estimated using diallel crosses. Analysis of variance for genotypes revealed highly significant differences (p=<0.01) for all traits. Components of combining ability analysis exhibited that, GCA was highly significant (p=<0.01) for 50% flowering, number of primary branches/plant and number of pods main/raceme, while non-significant for maturity and plant height whereas SCA and RCA effects were highly significant for all traits. GCA effects were higher than SCA and RCA for majority of the traits indicating that additive gene effects controlled the expression of these traits. RCA effects were greater than GCA and SCA for maturity showing that reciprocal crosses have higher potential than direct crosses for this trait. The parent genotypes NUR1, NUR2, NUR4, NUR7 and NUR8 and hybrids NUR1 x NUR3, NUR2 x NUR3, NUR2 x NUR4, NUR2 x NUR8, NUR4 x NUR2, NUR7 x NUR1, NUR8 x NUR2, NUR8 x NUR3 and NUR9 x NUR4 had higher GCA and SCA effects, respectively therefore could be used to develop early maturing, medium height and high yielding lines.
INTRODUCTION
Rapeseed-mustard is the second major source of edible oil after cottonseed in Pakistan and accounts for about 25% of the total production of edible oil of the country. Its oil has several advantages over other edible oils: it contains the least amount of harmful saturated fatty acids and also supplies two essential fatty acids viz., linoleic and linolenic, lacking in many other edible oils (Agnihotri and Kaushik, 2003). Majority of the local varieties contain low levels of oleic acid and high levels of erucic acid in the oil and of glucosinolates in the oil-free meal. Lower levels of erucic acid with a corresponding increase in oleic acid are preferred for nutrition, whereas low glucosinolate is preferred for feed. High levels of oleic acid make the oil stable and thus increase its shelf life. The cultivation of B. napus in the country is restricted due to late maturity and susceptibility of pods shatter. The efforts made by scientists to introduce exotic canola quality (commonly known as double low or OO) cultivars met with limited success due to their unsuitability to local agro-climatic conditions. Adding to the dilemma was the fact that erucic acid and glucosinolate contents are governed by multiple recessive genes. With this background, plant breeders took up the challenge to bring the local rapeseed-mustard to international standards of canola quality by developing new early maturing, high yielding containing canola quality B. napus lines through intraspecific/interspecific hybridization.
Advancement in the yield of brassica requires certain information regarding the nature of combining ability of parents available for use in the hybridization program. In addition, information about the nature of gene action involved in expression of quantitative and qualitative traits of economic importance is also required to develop desirable lines. Information on general and specific combining ability effects is very important in conducting a successful breeding program. Many workers have reported combining ability analysis (GCA and SCA) effects for different traits in brassica (Agnihotri et al., 1999; Kaushik and Agnihotri, 1999; Agnihotri and Kaushik, 2003; Abercrombie, et al., 2005; Satwinder et al., 2000; Kumar et al., 1997; Thakur and Bhateria, 1993; Diwakar and Singh, 1993). The main objective of the present study was to identify the best combiners and their crosses on the basis of their general, specific and reciprocal combining ability for maturity and some important traits.
MATERIALS AND METHODS
The study was conducted at the experimental farm of Department of Plant Breeding and Genetics, NWFP Agricultural University, Peshawar. Eight Brassica napus L. genotypes viz., NUR1, 2, 3, 4, 5,7,8 and NUR9 (NUR for National Uniform Rapeseed Yield Trail) were crossed manually in all possible combinations in a 8x8 diallel fashion during Rabi season 2004-05. In 2005-06, parents and their 56 F1 hybrids were grown in a randomized complete block design with three replications. All the F1 hybrids and their parent lines were randomly assigned to experimental plots. Each plot comprised four rows of 4 m each with 1 m space between rows and 15 cm between plants.
Data was recorded on 10 randomly selected plants from each parent and hybrid on 50% flowering, days to physiological maturity, plant height, number of primary branches/plant and number of pods main/raceme. The data collected were subjected to analysis of variance according to Steel and Torrie (1980). Combining ability effects were computed according to Griffing (1956).
RESULTS
Combining ability is the important tool for the evaluation of line (s) potential performance to be combined with another line (s). Analysis of variance presented in Table 1 showed that for genotypes highly significant differences (p≤0.01) were obtained for all traits studied viz., days to 50% flowering, days to physiological maturity, plant height, number of primary branches/plant and number of pods main/raceme. Since genotypic variations were significant for all the characters therefore these variations were partitioned into variation due to general combining ability (GCA; diagonal values), specific combining ability (SCA; above diagonal) and reciprocal combining ability (RCA; below diagonal) effects (Table 2). The results obtained are presented below.
| Table 1: | Mean squares for 50 and 100% flowering, maturity, plant height, branches/plant and pods main/raceme in Brassica napus L. |
| **= Highly significant (p≤0.01) | |
| Table 2: | Analysis of variance for combining ability of eight parents and their all possible F1s for 50 and 100% flowering, maturity, plant height, branches/plant and pods main/raceme in Brassica napus L. |
| ns = non-significant, ** = Highly significant | |
| Table 3: | General combining ability effects (diagonal values), specific combining ability effects (above diagonal) and reciprocal effects (below diagonal) for days to 50% flowering in Brassica napus |
| SE of parent values = 0.12, SE of F1 values = 0.15, F-values = 1.26, Variance of GCA = 2.385, Variance of SCA = 1.683 | |
| Table 4: | General combining ability effects (diagonal values), specific combining ability effects (above diagonal) and reciprocal effects (below diagonal) for days to physiological maturity in Brassica napus |
| SE of parental value = 0.14, SE of F1 value = 0.14, F-value (EP/EF1) = 1.07, Variance of GCA = 0.013, Variance of SCA = 0.17 | |
Days to 50% flowering: The mean squares due to GCA, SCA and RCA were highly significant (p≤0.01) for days to 50% flowering (Table 2). If flowering is early, seed filling period is sufficiently larger to have proper seed formation which ultimately results into bold seed. Therefore negative combining ability effects for days to flowering were desirable. According to Table 3, out of 8 parents, 4 showed negative GCA effects for 50% flowering. The maximum negative effects (-1.89) were noted in NUR1, while the maximum positive GCA effects (1.60) were recorded in NUR9. Out of 28 crosses 17 showed negative SCA effects. The best specific performance (SCA) with the maximum negative effects (-2.48) was presented by the cross NUR2xNUR3, while the maximum positive SCA effects were (3.26) found in cross NUR3xNUR8. Data regarding RCA showed that out of 28 crosses, 15 crosses showed negative effects for 50% flowering. The maximum negative (-3.50) reciprocal effects (RCA) were recorded in NUR8xNUR2 and NUR9xNUR4.
Days to physiological maturity: Table 2 shows that mean square values for GCA were non-significant while SCA and RCA were highly significantly (p≤0.01) different for days to physiological maturity. Since early maturity is one of the major objectives of brassica breeding programs therefore negative effects for GCA, SCA and RCA were preferred in these studies. Based on the data presented in Table 4, out of 8 parents 4 displayed negative GCA effects for days to physiological maturity where maximum negative effects (-0.21) were observed in NUR4, while the maximum positive GCA effects (0.15) were found in NUR7.
| Table 5: | General combining ability effects (diagonal values), specific combining ability effects (above diagonal) and reciprocal effects (below diagonal) for plant height in Brassica napus |
| SE of parental value = 0.86, SE for F1 values = 38.71, F-values (EP/EF1) = 44.91, Variance of GCA = 24.15 , Variance of SCA = 63.475 | |
| Table 6: | General combining ability effects (diagonal values), specific combining ability effects (above diagonal) and reciprocal effects (below diagonal) for number of primary branches/plant in Brassica napus |
| SE of parental value = 0.29, SE for F1 values = 0.36, F-values (EP/EF1) = 1.25, Variance of GCA = 0.79 , Variance of SCA = 1.173 | |
Regarding SCA effects, Table 4 showed that out of 28 crosses 14 showed negative SCA effects, where the best specific performance was observed in cross NUR2xNUR3 with the maximum negative value of -1.46. While the maximum positive SCA effects (0.71) were recorded in cross NUR1xNUR2. Data referring to RCA effects showed that out of 28 crosses negative reciprocal effects were maximum (-1.00) in NUR7xNUR1.
Plant height: Table 2 showed that mean square values for GCA were non-significant while SCA and RCA were highly significant (p≤0.01) for plant height. In brassica (grown for seed yield purposes) taller plants are susceptible to lodging thus medium/short statured plants are desirable therefore; negative GCA and SCA are suited for plant height. According to Table 5, out of 8 parents, 4 showed negative GCA effects for plant height, where the maximum negative effects (-5.70) were recorded in NUR2, while positive GCA effects were maximum (4.21) in NUR8. Out of 28 crosses 13 showed negative SCA effects (above diagonal values in Table 5) where the best specific performance was presented by the cross NUR2xNUR4 with the least value of -12.66. The maximum positive SCA effects (16.66) were found in cross NUR8xNUR9.
| Table 7: | General combining ability effects (diagonal values), specific combining ability effects (above diagonal) and reciprocal effects (below diagonal) for number of pods main/raceme in Brassica napus |
| SE of parental value = 2.94, SE for F1 values = 4.45, F-values (EP/EF1) = 1.51, Variance of GCA = 62.062, Variance of SCA = 41.218 | |
Data regarding RCA, indicated that out of 28 crosses, 14 crosses obtained negative values where the maximum negative reciprocal effects (-18.90) were in NUR8xNUR3.
Number of primary branches/plant: The mean squares presented in Table 2 showed that effects due to GCA, SCA and RCA were highly significant (p≤0.01) for number of primary branches/plant. Since number of primary branches/plant is directly linked with the final yield in brassica therefore, positive combining ability is considered suitable. According to Table 6, out of 8 parents 6 showed positive GCA effects for number of primary branches/plant, where the maximum effects (0.60) were recorded in NUR7 and NUR8, while negative GCA effects were maximum in NUR4 (-1.64). Out of 28 crosses, 14 crosses showed positive SCA effects, with the best specific performance (1.95) being observed in cross NUR1xNUR3, while negative SCA effects were maximum (-1.17) in cross NUR3xNUR9 and NUR4xNUR8. Out of 28 crosses, 16 crosses showed positive RCA effects where the maximum (2.50) figures were recorded in NUR7xNUR1.
Number of pods main/raceme: The mean squares referring to number of pods main raceme showed that effects due to GCA, SCA and RCA were highly significant (p≤0.01). According to Table 7, out of 8 parents 5 showed positive GCA effects for number pods main/raceme. The maximum GCA effects were observed in NUR7 (9.96), while negative GCA effects were maximum in NUR1 (-10.53). Data regarding SCA effects indicated that out of 28 crosses, 14 showed positive SCA effects where the best specific performance with the maximum positive SCA effect (12.84) was presented by the cross NUR2xNUR8, while negative SCA effects were maximum (-10.28) in cross NUR3 NUR9. Figures regarding RCA effects showed that out of 28 crosses, 10 crosses showed positive effects with the maximum (9.50) being observed in NUR4xNUR2.
DISCUSSION
Early flowering provides sufficient time for seed filling which could result in greater yield. Similarly early maturity and reduced plant height are desirable traits in brassica. Hence negative combining ability for these traits is required for better yields. In our experiment significant differences were recorded for all traits. GCA effects were higher for majority of the traits indicating the involvement of additive gene action. Our results are supported by the earlier findings of Fray et al. (1997) who reported that mean squares due to general and specific combining ability were significant for different traits in brassica. Components of combining ability showed that significantly negative effects were recorded both for parents and F1s for flowering, maturity and plant height indicating that these parents and crosses could be used for developing early maturing and short statured genotypes with higher yields. Our results are in agreement to the reports of Wang et al. (1997) who recorded significant GCA effects for flowering. In addition they found 14 hybrids with earlier flowering and 3 hybrids containing shorter plants. Present findings are similar to the reports of Teklewold and Becker (2005), who found significant Specific Combining Ability (SCA) effects for days to flowering, maturity, secondary branches and pods/plant. Krzymanski et al. (1997) revealed significant GCA or SCA effects for seed yield, oil content and some flowering traits.
For number of primary branches/plant and number of pods main/raceme positive effects were observed for combining ability components indicating that these parents and crosses could be used for developing high yielding genotypes. Similarly, Teklewold and Becker (2005) reported significant Specific Combining Ability (SCA) effects for yield associated traits. Significant combining ability effects for yield associated traits have also been reported by Kumar et al. (1997), Diwakar and Singh (1993) and Thakur and Bhateria (1993) in Brassica.