Variation and mean performance
Mean squares for the studied parents and their F
1 genotypes revealed highly significant variations for all characters (Table 1). That may indicate a wide genetic variability for studied characters, which may facilitate genetic improvement using such genetic pools of faba bean. Components of genetic variance, average degree of dominance, predictability ratio and heritability in narrow sense and genetic advance have been presented in Table 2.
The high estimates of genotypic and phenotypic coefficient of variation and high heritability in broad sense along with high genetic advance in per cent of mean were recorded for number of pods per plant. The high estimates of PCV and heritability with high genetic advance were recorded for grain yield per plant. The low estimates of PCV, GCV and high heritability with moderate genetic advance in percent of mean mentioned for five characters
viz. number of seeds per pod, harvest index, 100-seed weight, days to 50 per cent flowering and days to maturity indicated that these would be ideal traits for improvement through selection owing to existence of high heritability for them.
Combining ability analysis
The understanding of inheritance of various characters and identification of superior parents are important pre-requisites for launching an effective and efficient breeding programme. It is not always necessary that parents with high mean performance for yield and other traits would produce desirable F
1’s and/or segregants. The selection of a few parents having high genetic potential as per breeding objectives is essential because analysing and handling of very large number of crosses resulting from numerous parents available in germplasm collections of a crop would be an impractical and perhaps impossible task. Combining ability analysis is useful technique for understanding genetic worth of parents and their crosses for further exploitation in breeding programmes.
In addition, it also provides the idea about gene effects involved in inheritance of different characters which is essential for deciding suitable breeding strategy. Among the various techniques of combining ability analysis, line × tester analysis (
Kempthorne, 1957) has been widely utilized for screening of germplasm to identify valuable donor parents and promising crosses in many crops including faba bean
(Abdelmula, 2006). The present study, therefore, aims to study the combining ability of parents and crosses and gene action for grain yield and its components using line × tester technique. The important findings of the analysis are discussed below:
Gene action and components of genetic variance
The analysis of variance for combining ability for twelve characters have also been presented in Table 1a, while estimates of components of genetic variance and other genetic parameters are given in Table 2. The mean squares due to replications appeared non- significant for all the traits. The mean squares due to testers emerged non-significant for all the characters under study. The variance due to lines was found to be non-significant for all the characters except plant height, number of pods per plant, biological yield per plant and grain yield per plant. The mean squares due to lines × testers interactions, representing importance of specific combining ability and non-additive gene effects, were found to be highly significant for all the twelve characters under study. The above discussion suggests predominant role of non-additive gene effects represented by specific combining ability variances for all the characters.
The estimates of
SCA variance were higher than the corresponding estimates of
GCA variance for all the traits. The values of average degree of dominance were more than unity (>1) revealing over dominance for all the characters except plant height. The predictability ratio was less than one for all the characters indicated the preponderance of non-additive gene action. Earlier workers
i.e. Obiadalla-Ali et al., (2013) reported that
SCA variances were greater for number of branches per plant, seed yield and number of pods per plant which indicated the preponderance of non-additive gene action.
Ibrahim (2010) also reported non-additive gene action for seed yield per plant.
Haridy and Amein (2011) got the same results for number of pods per plant and pod length.
The estimates of heritability in narrow sense (h
2ns) have been classified by
Robinson (1966) into three categories
viz., high (>30%), medium (<30 - >10) and low (<10). High estimates of heritability in narrow sense were found for plant height, number of pods per plant, grain yield per plant, harvest index and biological yield per plant, while medium estimate of heritability in narrow sense were recorded for pod length, number of branches per plant, days to maturity, number of seeds per pod, 100-seeds weight and days to 50 per cent flowering. The low estimate of heritability was recorded for protein content. Low estimates of genetic advance were recorded for all the characters studied.
The little role of additive gene effects of fixable nature for grain yield and most of other yield components in the present study suggested that these traits are not amenable to improvement through selection even in early generations. This indicated that considerable improvement in status of grain yield and important yield attributes in faba bean can not be achieved by following conventional breeding procedures normally used in often cross pollinated crops leading to development of inbreds lines. The predominance of non-additive gene effects representing non-fixable dominance genetic variance indicated that maintenance of heterozygosity would be highly fruitful for improving the characters. Hence, the suitable breeding strategy for attaining high yield would be the full or partial exploitation of heterosis through development of hybrid, synthetic or composite cultivars. Since, the technology for development of faba bean hybrids for commercial purposes is being widely and successfully used in different countries, it is recommended to explore possibility of isolating high yielding commercial hybrids utilizing the materials of the present investigation.
General combining ability
For illustrating genetic worth of parents for hybridization programme, the general combining ability (GCA) effects of eighteen parents (fifteen lines + three testers) for twelve characters are consolidated in Table 3. The significant and positive
GCA effects for grain yield per plant were exhibited by four lines, HB 10 (3.02), HB 50 (2.88), EC 454751 (2.54) and EC 301470 (1.80). The parent with highest
GCA effects for grain yield, HB 10, also showed significant
GCA effects in desirable direction for plant height, number of branches per plant, number of pods per plant, biological yield per plant and 100-seed weight. Testers, HB 09-16 and HB 09-15 showed good general combining ability for grain yield per plant. Tester, HB 09-16 also exhibited significant and positive
GCA effect for plant height, pod length, number of seeds per pod, harvest index and 100-seed weight.
The six parents
viz. HB 10 (3.02), HB 50 (2.88), EC 454751 (2.54) and EC 301470 (1.80), HB 09-16 (0.59) and HB 09-15 (0.52) showing positive and significant
GCA effects for grain yield and other important traits as mentioned above may serve as valuable parents for hybridization programme or multiple crossing programme for obtaining high yielding hybrid varieties or transgressive segregants for developing high yielding varieties of faba bean. Other lines identified as good general combiners in desirable direction for characters other than grain yield per plant may also be recommended for exploitation in hybridization programme as donor of component characters for which they emerged as good general combiner in spite of being average or poor general combiner for grain yield.
It is evident that most of the lines showing significant positive
GCA effects for grain yield per plant also exhibited positive and significant
GCA effects for some of the important yield components traits. This indicated that the significant
GCA effects for grain yield in positive direction resulted from similar
GCA effects of some other yield components suggesting that the combining ability for grain yield was influenced by the combining ability of its components. Therefore, simultaneous improvement in important yield components and other associated traits alongwith grain yield may be better approach for raising yield potential in faba bean.
Specific combining ability effects
The specific combining ability (
SCA) effects, which are supposed to be manifestation of non-additive components of genetic variance, are highly valuable for discrimination of crosses for their genetic worth as breeding materials. The estimates of
SCA effects of forty five crosses for twelve characters are given in Table 4.
Out of forty five crosses, eleven emerged with positive and significant
SCA effects for grain yield per plant. The top ten crosses EC 243626 × HB 09-16, HB 30 × HB 09-15, EC 329627 × HB 09-15, EC 263620 × HB 09-16, EC 301470 × DFB 14-, IC 10845 × HB 09-16, EC 454751 × DFB 14-1, EC 25085 × DFB 14-1, IC 588958 × HB 9-16 and EC 329706 × HB 9-16 showed significant and positive
SCA effects for grain yield per plant as well as some other yield components. The cross having highest positive and significant
SCA effects for grain yield per plant, EC 243626 × HB 09-16 also recorded significant
SCA effects in desirable direction for days to 50 per cent flowering, days to maturity, plant height, number of branches per plant, number of pods per plant, pod length, biological yield per plant and harvest index. The second ranking cross for higher positive and significant
SCA effect for grain yield per plant, HB 30 × HB 09-15 showed significant and desirable
SCA effects for number of pods per plant, pod length, number of seeds per pod, biological yield per plant, harvest index and 100-seed weight. Similarly, remaining nine crosses having significant and positive
SCA effects for grain yield per plant also possessed significant
SCA effects in desirable direction for some other characters also.
The eleven crosses having positive and desirable
SCA effects for grain yield and some of its component traits merit attention in breeding programme for exploitation as hybrid cultivars. The eleven crosses having significant and positive
SCA effects for grain yield per plant also showed positive and desirable significant
SCA effects for some other characters, most commonly biological yield per plant, 100-seed weight, pod length, plant height and days to 50 per cent flowering. This suggested that manifestation of
SCA effects for grain yield is related with higher
SCA effects for important yield components.
In general, the crosses showing significant and desirable
SCA effects were associated with better
per se performance for respective traits. However, the crosses having high
SCA effects in desirable direction did not always have high mean performance for the character in question. Thus, the
SCA effect of the crosses may not be directly related to their
per se performance. This may be attributed to the fact that
per se performance is a realized value whereas
SCA effect is an estimate of F
1 performance over parental one. Therefore, both
per se performance along with
SCA effects should be considered for evaluating the superiority of a cross although the former may be more important if development of F
1 hybrids is the ultimate objective.
The crosses exhibiting high order of significant and desirable
SCA effects for different characters involved parents having all types of combinations of
GCA effects such as high × high (H × H), high × average (H × A), high × low (H × L), average × average (A × A), average × low (A × L) and low × low (L × L) general combiner parents. The foregoing observation clearly indicated that there was no particular relationship between positive and significant
SCA effects of crosses with
GCA effects of their parents for the characters under study.
Heterotic response
The heterosis breeding has been used extensively in improving yield potential through development of hybrid cultivars in most of the often cross pollinated crops like faba bean. The exploitation of heterosis for developing high yielding commercial hybrids for faba bean has been found highly fruitful. The presence of high heterosis for economically important characters is not only useful for developing hybrids, synthetic or composites through exploitation of heterosis, but also helps in obtaining transgressive segregants for development of superior inbred lines. In present study, the estimates of heterosis over better-parent and standard varieties, SV
1 (HFB 1) and SV
2 (Vikrant) were calculated for forty five F
1’s to assess their genetic potential as breeding material.
A wide range of variation in the estimates of heterobeltiosis and standard heterosis in positive and negative direction was observed for grain yield per plant (Table 5). In case of grain yield per plant, heterobeltiosis ranged from -35.75 to 100.74 per cent and standard heterosis varied from -35.10 to 46.30 per cent over SV
1 (HFB 1) and from -33.26 to 50.46 per cent over SV
2 (Vikrant).
Sixteen crosses exhibited positive and significant heterosis over BP, nineteen crosses exhibited positive and significant heterosis over both SV
1 and SV
2. The best ten cross combinations were HB 30 × HB 09-15 (100.74%), EC 243626 × HB 09-16 (63.82%), HB 10 × HB 09-16 (61.69%), HB 10 × HB 09-15 (61.30%), EC 454751 × DFB 14-1 (43.09%), EC 454751 × HB 09-16 (40.72%), HB 30 × DFB 14-1 (39.49 %), HB 10 × DFB 14-1 (38.54%), EC 329706 × HB 09-16 (37.37%), EC 454751 × HB 09-15 (28.04 %) over BP. The top five crosses for standard heterosis over SV1 were EC 243626 × HB 09-16 (46.30%), HB 30 × HB 09-15 (45.12%), HB 50 × HB 09-15 (41.28%), HB 10 × HB 09-16 (40.29%) and HB 10 × HB 09-15 (39.95%). The best five cross combination over SV
2 were EC 243626 × HB 09-16 (50.46%), HB 30 × HB 09-15 (49.25%), HB 50 × HB 09-15 (45.30%), HB 10 × HB 09-16 (44.28%) and HB 10 × HB 09-15 (43.93%).
Besides, grain yield, substantial heterosis over better-parent and standard varieties was also observed in negative as well as positive direction for different characters (Table 5). However, the number of crosses showing significant estimates and the range of heterosis varied from one character to another. The mean heterosis was both in positive and negative direction for different characters. In general, some crosses showed appreciable and high heterosis for most of the characters under study. The existence of wide spectrum of heterosis in either direction with expression of high degree of desirable heterosis by some crosses for most of the characters observed in present study is in conformity with the earlier reports for such characters in faba bean (
Alghamdi, 2009;
Ibrahim, 2010;
Mourad, 2011;
Farag and Afiah, 2012;
Abd-El rahman et al., 2012;
Bakhit and Abdel-Fatah, 2013;
Obiadalla-Ali et al., 2013; Zeinab and El-Emam, 2013;
El-Banna et al., 2014; Zeinab and Helal, 2014;
Ashrei et al., 2014; Bishnoi et al., 2015; 2017).
It was also noted that higher heterosis over better-parent was found in some lower yielding crosses when compared to other crosses which have displayed high yield. This suggested that while selecting the best hybrid, besides the heterotic response over better-parent, the mean performance of the crosses should also be given due consideration. Since, heterosis estimate results from F
1-BP and depends more or less on the mean of the parents in question, there is every possibility of getting a cross with lower mean performance but high heterotic response, in case the parental performance is very poor. On the contrary, there can be a cross with high mean performance but low heterosis in case parental performance is also high. The mean performance being the realized value and the heterotic response being an estimate, the former should be given due consideration while making selection of cross combinations especially when objective is to identify a hybrid for commercial cultivation as in present case.