Combining Ability Studies for Development of New Hybrids over Environments in Sunflower ( Helianthus annuus L . )

Seven CMS lines were crossed with six inbred lines in Line x Tester fashion to elucidate the information on the nature of gene action involved in the inheritance of important quantitative traits and to select the parents with good gca and crosses with good sca effects. The resultant 42 hybrids were evaluated along with their parents with three standard checks at three locations in Andhra Pradesh state viz., Hyderabad, Tandur, and Jagtial. The pooled analysis of variance for combining ability revealed that sca variance was higher in magnitude compared to gca variance for all the characters except oil content indicating the preponderance of non-additive gene action for all the characters while additive gene action for oil content. The gca effects of the parents in pooled analysis revealed that among the lines ARM 243B and CMS 17B and among the testers RHA-6D-1R, RES-834-1 and 3376R were found to be promising general combiners for seed yield and yield component characters. Based on significant sca effects in pooled analysis, five hybrids viz., CMS 89A x RES-834-1, CMS 17A x LTRR 341, ARM 243A x R 298, ARM 238A x 3376R and CMS 852A x R-649 were identified as promising for seed yield and other yield contributing characters.


Introduction
Breeding research for genetic enhancement in sunflower has been in progress for more than 3 decades in India.The objective of sunflower breeding is to develop the high yielding hybrids with high oil quality or disease resistance Dudhe et al. (2009).Line x Tester analysis provides a systematic approach for detection of appropriate parents and crosses in terms of investigated traits (Kempthorne, 1957).Combining ability studies elucidate the nature and magnitude of gene action involved in the inheritance of character by providing the information on the two components of variance viz., additive genetic and dominance variance, which are important to decide upon the parents and crosses to be selected for eventual success.Paying due consideration to genotype x environment interaction during the studies on combing ability may be helpful in identifying desirable genotypes and in understanding the precise nature of inheritance of economic traits.Accordingly, the present investigation was undertaken to have an idea on the nature of gene action involved in the inheritance of important quantitative traits and to select the parents with good gca and crosses with good sca effects through line x tester analysis over locations in sunflower.

Material and methods
Seven CMS lines viz., CMS 234A, CMS 17A, CMS 852A, CMS 89A, ARM 238A, ARM 243A and ARM 248A and six inbred lines viz., RES-834-1, LTRR 341, RHA-6D-1R, 3376R, R-298 and R-649 were planted during kharif, 2008 at Directorate of Oilseeds Research, Rajendranagar, Hyderabad and crossing was performed in line x tester fashion to produce 42 hybrids.During rabi, 2008-09 the 42 hybrids along with their parents and three standard checks viz., KBSH-1, PAC-1091 and KBSH-44 were evaluated in a Randomized Block Design replicated thrice at three different locations viz., Directorate of Oilseeds Research, Hyderabad; Agricultural Research Station, Tandur; and Regional Agricultural Research Station, Jagtial.Observations were recorded on five randomly selected plants in each hybrid combination per replication for ten quantitative characters.Data obtained were subjected to line x tester analysis (Kempthorne, 1957) to estimate general and specific combining ability effects and their respective variances.The result of pooled analysis over locations is presented.

Results and discussion
The pooled analysis of variance revealed significant differences due to environments for all the characters indicating the sufficient diversity among the environments (Table 1).Significant differences for replication x locations were recorded for plant height, number of filled seeds per head, seed yield per plant and oil yield per plant.The differences among the parents, parents vs crosses and crosses were observed to be significant for all the characters studied indicating the existence of wider genetic differences among parents and crosses.Partitioning of crosses into lines, testers and lines x testers revealed that the variance due to lines were significant for all the characters except days to maturity, number of filled seeds per head, number of unfilled seeds per head, seed yield per plant and oil yield per plant, whereas for testers, plant height, head diameter, number of unfilled seeds per head, seed yield per plant, oil content and oil yield per plant were found significant, indicates wide variability existing among the genotypes.The interaction due to lines x testers were significant for all the traits studied, suggesting that significant contribution of sca effects towards the variation among the crosses.The results are in accordance with the earlier studies Madhavilatha et al. (2004), Vishwanath and Shankar Goud (2006) and Gouri Shankar et al. (2007).Interaction effects of parents vs crosses x locations, parents x locations and crosses x locations were significant for all the characters, except head diameter, number of filled seeds per head and number of unfilled seeds per head in case of (parents vs crosses) x locations and oil yield per plant in case of parents vs locations.Further partitioning of crosses x locations indicated that the interaction of lines x locations showed significant differences for plant height, stem diameter and oil content, while testers x locations was significant only for oil content, suggesting the sensitivity of gca effects of parents to environmental fluctuations for these characters.Interaction effects of lines x testers x locations were significant for all the characters studied, indicates the sca effects of hybrids interacted with the environments for all the characters studied which are contradictory with the earlier reports of Sanket Sharma et al. (2003) and Binodh et al. (2008).The comparative estimates of variances due to general combining ability (gca) and specific combining ability (sca) revealed the predominance of sca variance in relation to gca variance for all the traits except oil content, which implied that all the characters were predominantly under the control of non additive gene action, while additive gene action for oil content.The results corroborates with the findings of Vishwanath and Shankar Goud (2006) and Gouri Shankar et al. (2007).Degree of dominance observed to be more than unity for seed yield and other yield components except oil content under study suggesting over-dominance is the major factor for high level of expression of these traits.These results are in accordance with the earlier reports of Satyanarayana (2000) and Madhavilatha et al. (2004).
The gca effects of the parents revealed that the lines ARM 243B and ARM 248B were good general combiners for most of the traits like head diameter, 100 seed weight, seed yield per plant, oil content and oil yield per plant by exhibiting significant positive gca effects (Table 2).These lines also showed negative significant gca effects for the traits like days to 50 per cent flowering and days to maturity indicating their usefulness in breeding for early maturing hybrids (Table 4).The line ARM 238B was best general combiner for number of filled seeds per head, head diameter, seed yield per plant, oil content and oil yield per plant.On the contrary, CMS 17B characterized by late maturity by also possessed favourable alleles for tallness, filled seeds per head, 100 seed weight, seed yield per plant and oil yield by recording gca effects in the positive direction.The line CMS 89B contributed a large number of favourable alleles for plant height (dwarf stature) and oil content as indicated by their high gca effects for these characters.Among the testers, RHA-6D-1R was best general combiner for majority of the important yield components, i.e., head diameter, number of filled seeds per head, 100 seed weight, seed yield per plant, oil content and oil yield.It also exhibited high negative gca effect in a desired direction for unfilled seeds and days to maturity.The tester LTRR-341 possessed favourable genes for early flowering, maturity and dwarf plant type by recording significant negative gca effects.Favourable genes for oil content were predominantly contributed by RES-834-1, RHA-6D-1R, R 649 and R 298 by recording significant positive gca effects.For the traits, days to maturity, number of filled seeds per head, seed yield, oil content and oil yield, the tester RES-834-1 was identified as good combiner, as it exhibited significant positive gca effects, The line 3376R was a good combiner for number of filled seeds per head, seed yield and oil yield.Hence, among the lines ARM 243B and CMS 17B and among the testers RHA-6D-1R, RES-834-1 and 3376R were proved to be good combiners for seed yield and its most of the related characters and need to be exploited in future breeding programme.The parents which are good general combiners for yield possessed gca effects in the desired direction for yield components was also reported earlier by Radhika (1994).
Sixteen crosses were identified as good specific combinations for seed yield per plant (Table 3).Among these crosses the positive sca effect for seed yield also exhibited significant positive sca effects for other yield contributing traits.The cross combination CMS 17A x LTRR 341 showed significant positive sca effects for seed yield per plant along with number of filled seeds per head, 100 seed weight and oil yield per plant along with negative significant sca effect for days to 50% flowering and days to maturity indicating its suitability for commercial exploitation for early maturing hybrids (Table 5).In the crosses viz., ARM 238A x RES-834-1, ARM 238A x 3376R and ARM 243A x 3376R, involved parents with high x high gca effects suggesting additive x additive type of gene action.This indicated that selection could be effective in F 2 generation and utilized in transgressive breeding.Similar findings as observed in present study were also reported by Sanket Sharma et al. (2003), Manivannan et al. (2005) and Binodh et al. (2008).On the other hand crosses CMS 17A x LTRR 341, ARM 243A x R 298, ARM 248A x R 649 and CMS 89A x RES-834-1 involved parents with high x low gca effects indicating the involvement of additive x dominance genetic interaction.Peng and Virmani (1990) also reported about the possibility of interaction between positive alleles from good combiner and negative alleles from poor combiners in high x low crosses in sunflower.The cross combinations CMS 89A x R 649, CMS 852 x LTRR-341 and CMS 852A x R-649 involved low x low combining parents indicating over dominance and epistatic interactions.The results clearly indicated that the high performance of hybrid need not be the ones with high sca effect and vice-versa.Similar results were also obtained by Vishwanath and Shankar Goud (2006) and Gouri Shankar et al. (2007).In spite of the involvement of both poor general combiners in some crosses or one of the parents as poor general combiner, these cross combinations expressed significant sca effects in desirable direction which might be due to concentrations and interaction between favourable genes contributed by parents.
The crosses namely, CMS 89A x LTRR 341 and CMS 852A x R-649 and ARM 238A x R-298 were the best specific combiners for oil content having high x low, low x high and high x high gca parental combination, indicating a genetic interaction of the additive and non additive types.Based on significant high sca effects, five hybrids viz., CMS 89A x RES-834-1, CMS 17A x LTRR 341, ARM 243A x R 298, ARM 238A x 3376R and CMS 852A x R-649 were identified as promising for seed yield and other yield contributing characters.The performance of these crosses needs to be critically evaluated over different seasons and locations to confirm their superiority and stability.

Table 1 .
Pooled analysis of variance for combining ability (L x T) for yield and yield components in sunflower

Table 2 .
Estimates of general combining ability effects for thirteen parents for ten characters in sunflower

Table 3 .
Estimates of specific combining ability effects for forty two crosses for ten characters in sunflower