Studies on Combining Ability for Seed Yield and Its Related Traits in Blackgram [Vigna mungo (L.) Hepper]

Blackgram [Vigna mungo (L.) Hepper], 2n=2x=22 popularly known as urdbean, black lentil, mungobean, matpe bean or mash, belonging to the sub-genus Ceratotropis, is a grain legume, originated from central Asia and India with Vigna mungo var. silvestris as its progenitor which is found wild in India (Lukoki et al., 1980). It is one of the nutritious pulse crops, extensively cultivated in the Indian sub-continent and to a lesser extent in Thailand, Australia and other Asian and South Pacific countries (Poehlman 1991). It is an important short duration self pollinated pulse crop grown in many parts of India. It forms a significant component of dietary carbohydrates (60%), fat (1.5%), minerals, amino acids and vitamins. Compared to other legume requirement in Indian sub continent and is highly nutritious with protein (25-26%), crops, urdbean is known for its synchronous maturity, non-shattering pods, more clusters and pods with large seeds. However, its yield is low in contrast to other grain legumes. Low productivity in this crop is attributable to its narrow genetic base due to common ancestry of various superior genotypes, poor plant type, cultivation under marginal and harsh environmental conditions and vulnerability to abiotic and biotic stresses. Hence, there is a strong need to improve the productivity of blackgram. This could be achieved by studying the genetic architecture of this crop. However, success depends primarily upon identification of best parental lines which may produce desirable gene combinations. Selection of parental lines on the basis of their genetic value is key requirement for any successful crop improvement programme.
       
Combining ability is the relative ability of a genotype to transmit its desirable performance to its crosses. It is one of the powerful tools available to estimate the combining ability effects and it aids in selecting the desirable parents and crosses for the exploitation of heterosis (Sarker et al., 2002; Muhammad et al., 2007). The knowledge of the type of gene action involved in the expression of yield and component traits is essential to choose an appropriate breeding strategy to isolate desirable segregations in later generation. The present investigation was undertaken to generate information on the nature of gene action and combining ability in line × tester biometrical technique to identify elite parents and superior hybrid combination in blackgram.

Twenty-nine genetically diverse genotypes of blackgram (Vigna mungo L.) obtained as Indigenous collection from IIPR, Kanpur were crossed in line × tester mating design. Twenty-seven (Lines) genotypes were used as a female (IC-281980, IC-282007, IC-282008, IC-436910, IC-281989, IC-398973, IC-281982, IC-282001, IC-398956, IC-413306, IC-413307, IC-281993, IC-343962, IC-413305, IC-398998, IC-281992, IC-413304, IC-436852, IC-282002, IC-281990, IC-281995, IC-343885, IC-343947, IC-282004, IC-413309, IC-343943 and IC-281984) and two (Testers) as male parents (HPBU-111 and Him Mash-1) to produce 54 hybrid combinations during Kharif 2016. These F₁ hybrids along with their parents were evaluated in randomized block design (RBD) with three replicates during Kharif 2017 at the Experimental Farm, Department of Crop Improvement, CSK HPKV, Palampur (HP).
       
The data were recorded on ten random competitive plants in each replication for the traits viz., plant height (cm), branches per plant, pods per plant, pod length (cm), seeds per pod, 100-seed weight (g), biological yield per plant (g), seed yield per plant (g) and harvest index (%). The protein content (%) was determined by using macro-Kjeldhal method Jackson (1976). For, days to 50% flowering and days to 75% maturity observations were recorded on plot basis. Recommended package of practices were followed for raising the crop. The data so generated were analyzed as per Panse and Sukhatme (1984). Combining ability analysis was carried out following the method of Kempthorne (1957). Estimates of additive and dominance variance, average degree of dominance and narrow sense heritability were calculated as per line × tester analysis.

Analysis of variance for parents (27 lines and two testers) and their hybrids revealed highly significant genotypic differences for all the traits namely days to 50%  flowering, days to 75% maturity, plant height (cm), branches per plant, pod length (cm), pods per plant, seeds per pod, seed yield per plant (g), biological yield per plant (g), 100-seed weight, harvest index (%) and protein content (%) indicating substantial amount of genetic variability in the material under study (Table 1). Hence, it offers a better scope for further improvement of breeding material by the selection of promising genotypes in blackgram breeding programme. The significant variation for all the traits was also observed by earlier workers viz., Kumar et al., (2017) for seed yield, number of branches, pods, clusters, 100-seed weight, fodder biomass and harvest index; Chauhan et al., (2018) for all studied traits; Bandi et al., (2018) for days to 50% flowering, days to maturity, plant height (cm), number of branches per plant, number of clusters per plant, number of pods per plant, pod length (cm), number of seeds per pod, 100-seed weight (g) and grain yield per plant (g) in blackgram.
 

 
Analysis of variance for combining ability
 
Analysis of variance for combining ability (Table 2) revealed that mean squares due to crosses were highly significant (P≤ 0.01) for all the traits under study. Similar results were observed by Patial et al., (2018). One should proceed for line × tester analysis only if the crosses mean squares are significant. The further partitioning of mean squares into lines, testers and line × tester interactions revealed that mean squares due to lines were highly significant (P≤0.01) for all the traits viz., days to 50% flowering, days to 75% maturity, plant height (cm), branches per plant, pod length (cm), pods per plant, seeds per pod, seed yield per plant (g), biological yield per plant (g), 100-seed weight, harvest index (%) and protein content (%). The mean squares due to testers were significant for days to 50% flowering, days to 75% maturity, branches per plant, pod length (cm), pods per plant, seeds per pod, seeds yield per plant (g), biological yield per plant (g) and harvest index (%). Highly significant mean squares for lines and testers indicated existence of additive variance. The lines exhibited greater magnitude of mean squares as compared to testers for all the traits except, for three traits viz., days to 50% flowering, days to 75% maturity and pod length. This indicated a wider genetic diversity of lines as compared to the testers for these traits. This observation was in agreement with the findings of Chakraborty et al., (2010) and Surashe et al., (2017).
 

 
Genetic parameters and proportional contribution of lines, testers and their interactions
 
In a breeding programme, once the appropriate parents and potential crosses are identified, the next important step is to adopt a suitable breeding strategy for the management of generated variability which largely depends upon type of gene action in the population for the traits under genetic improvement (Cockerham 1961; Sprauge 1966). The estimates of additive variances (σ2A), dominance variances (σ²D) and proportional contribution of lines, testers and their interactions to the total variances are presented in Table 3. Magnitude of dominance variance (σ²D) was found to be higher than additive variance (σ²A) for all the traits, which indicated the preponderance of non-additive gene action for these traits. Gill et al., (2014) found that the relative estimates of dominance variance were higher than additive variance for days to 50% flowering, days to maturity, plant height and number of pods per plant. The average degree of dominance more than unity reveal over dominance confirming the above findings. Hence, non-additive genetic variance was important in controlling these traits. Similar results were reported by Govindaraj and Subramanian (2001), Manivannan (2002), Selvam and Elangaimannan (2010), Chakraborty et al., (2010) and Thamodharan et al., (2017) for plant height, branches per plant, days to 50% flowering, days to maturity, cluster per plant, pods per cluster, 100-seed weight and pod length. Narrow sense heritability is the proportion of additive genetic variance to total phenotypic variance. The narrow sense heritability (h²ns) estimates were classified as high (>50%), medium (30-50%) and low (<30%). All the traits showed low narrow-sense heritability, indicating that non-fixable component of variation is governing by these traits (Table 3).
 

       
The proportional contribution of lines ranged from 34.31% (days to 50% flowering) to 74.34% (seed yield per plant). The proportional contribution of testers ranged from 0.33% (biological yield per plant) to 9.63% (days to 75% maturity) (Table 3). The proportional contribution of line × tester interactions ranged from 22.31% (seed yield per plant) to 62.65% (100 seed weight). The contribution of lines was found to be higher than individual contribution of testers for all the traits studied. This indicated the presence of wide genetic diversity among the lines as compared to the testers. The contribution of line × tester interaction was found to be higher than the individual contribution of testers for all the traits studied, indicating the higher estimates of variances due to specific combining ability which further confirmed the active involvement of non-additive gene action in the inheritance of different traits studied. The per cent contributions of lines, testers and their interactions to the total variance of various quantitative traits in the blackgram crosses were estimated by Chakraborty et al., (2010).
 
General combining ability (GCA) and specific combining ability (SCA)
 
The GCA and SCA effects of lines and testers are presented in Table 4 and 5. The GCA effects is a good estimates of additive gene action (Sprague and Tatum, 1942) reflecting the performance of parental lines in combination with all other lines. Estimates of the general combining ability (GCA) effect (Table 4) for the lines and testers for the traits indicated that no single parent was a good general combiner for all traits under study. However, line IC-436910 (11.55**), IC-281989 (7.71**), IC-398973 (4.86**), IC-413307 (2.11**), IC-413306 (1.44**), IC-343885 (1.34**), IC-436852 (0.93**) and IC-413305 (0.88**) were found to be good general combiners for seed yield per plant. The parental lines IC-436910 (2.30**), IC-343943 (2.27**), IC-398973 (2.24**), IC-343885 (2.11**), IC-413306 (1.99**) and IC-281995 (1.05*) were best general combiners for protein content.
 

 

       
Earliness is an important breeding objective. Because negative heterosis is required to achieve earliness, lines IC-436852 (-3.45**, -3.06**), IC-343947 (-2.62**, -2.72**), IC-282008 (-3.12**, -3.22*) exhibited significant GCA values for days to 50% flowering and days to 75% maturity respectively. Among testers, Him Mash-1 was good general combiner for earliness whereas, HPBU-111 was good general combiner for branches per plant, pod length (cm), pods per plant, seeds per pod, biological yield per plant (g), seed yield per plant (g) and harvest index (%). Overall, IC-436910, IC-413306, IC-398973 and IC-343885 lines were good general combiners for seed yield and other quantitative traits and among these lines, IC-436910 was the best general combiner for maximum number of traits i.e. days to 75% maturity, plant height, branches per plant, pod length, pods per plant, seeds per pod, biological yield per plant, seed yield per plant and protein content. The GCA effect revealed that in order to synthesize a dynamic population with most of the favorable genes, it will be pertinent to make use of these parents, which are good general combiner for several traits, in a multiple crossing program to obtain desirable transgressive segregants for higher seed yield.
       
Another important aspect from practical point of view, which needs consideration, is the identification of potential cross combinations for obtaining desirable transgressive segregants with a higher degree of probability. The specific combining ability is the deviation from the performance predicted on the basis of general combining ability (Allard, 1960). The cross combinations with highest SCA effects along with mean performances and GCA effects of parents are listed in Table 5.
       
The cross combination IC-436910 × HPBU-111 recorded significant positive SCA effects and high per se performance for days to 75% maturity, plant height, branches per plant and seed yield per plant. The cross IC-281980 × Him Mash-1 recorded significant positive SCA effects for plant height, branches per plant and biological yield per plant. IC-343947 × Him Mash-1 showed significant positive SCA effects for pod length and seeds per pod. For days to 50% flowering and 75% maturity, IC-281995 × Him Mash-1 was found to be highly significant negative SCA effects as well as low mean values indicating early maturity. It was observed that the desirable cross combination included high × high, medium × high, high × low, medium × medium and low × low type of general combiner (Table 5). The high × medium or vice-versa combination could be due to additive and additive × additive types of gene action which are fixable in nature. The desirable performance of cross combination like low × medium, medium × medium general combiners may be ascribed to complimentary gene effects.    

Various workers have recorded good general combiners and specific combiners for various traits in blackgram with different genetic material Gill et al., (2014), Kumar et al., (2014), Gill et al., (2015), Sanjeev et al., (2015), Balouria et al., (2016) and Thamodharan et al., (2017). Similarly, Barad et al., (2008), Patil et al., (2011), Nath et al., (2018) reported in mungbean. Hence, the lines and cross-combinations with significant and positive GCA, SCA effects could be utilized for selection of superior segregants for developing improved cultivars in future.

Estimate of genetic parameters in the present study described that the magnitude of dominant variance (σ²D) was found to be higher than additive variance (σ2A) for all the traits, which indicated the preponderance of non-additive gene action. Combining ability results showed that, the relative size of GCA variances was higher than SCA variances for most of the traits. Hence, is clear that quantitative traits governs yield were under the control of both additive and non-additive gene action.  It is concluded that breeding methods such as recurrent selection or biparental mating followed by selection would be ideal to exploit both non­-additive and additive gene action.