Combining Ability Studies for Seed Yield and its Quantitative Traits in F₂ Crosses of Mungbean [Vigna radiata (Wilczek)]

Mungbean [Vigna radiata (L.) Wilczek] belongs to family Fabaceae with chromosome number 2n=22. Pulses have advantages of sustaining different cropping system and fit for multiple cropping due to its characteristic feature to fix atmospheric nitrogen in association with rhizobium which in turn enriches the soil. Its increasing demand among people is due to its excellent source of digestible protein which shows low flatulence and it is complementary staple food among the low income groups. In India, pulses occupy 15% of the gross cropped area, out of which mungbean constitute only 2% of the total pulse area.
               
Low productivity of pulses in general and mungbean in particular is due to their cultivation in marginal land under rainfed condition, lack of high yielding genotypes, low yield potential, narrow genetic base, lack of yield stability and susceptibility to pests and diseases. Therefore, there is a compelling demand for developing high yielding varieties which can be accomplished by recombination breeding followed by selection in segregating generations in autogamous crop like mungbean. The pre-requisite of choosing the desirable parents for crossing programme can be done by combining ability analysis. On that account, diallel method is one of the most efficient method for determining combining ability which estimates the general combining ability (GCA) and specific combining ability (SCA). GCA is the result of additive × additive gene action whereas, SCA is result of dominance, additive × additive and dominance × dominance gene action. The nature of gene action would determine both the rate of advance and the ultimate gain from selection, thus, helping in planning a sound breeding programme.

The basic experimental material comprised of six improved varieties of mungbean which included three improved varieties of OUAT, Odisha (Kamdev, OBGG-52 and OBGG-58), one each from PAU, Ludhiana (MGG-385), GBPUAT, Pantnagar (Pant Mung-8) and IIPR, Kanpur (IPM-02-14). Fifteen crosses were developed in half diallel fashion by growing six parental varieties at staggered sowing starting from September, 2018 to October, 2018. In summer 2018-19, the parents and F₁s were grown in EB-II, Experimental field of OUAT, Bhubaneswar to raise parental and F₂ cross seeds. During Rabi season of 2019-20, six parents and 15 F₂ crosses were evaluated in the same field in a randomized block design with three replications. They were grown in 3 rows of 3m row length with spacing of 30×10 cm. Each entry was harvested replication wise when the green pods turned to blackish - brown colour. Sampling was done using 30 individual plants for each entry and each replication and observations were recorded for the ten different characters viz., days to 50% flowering, days to maturity, plant height (cm), number of branch/plant, number of cluster/plant, number of pods/plant, pod length (cm), number of seeds/pod, seed yield/plant (gm) and 100-seeds weight (gm) from each entry of each replication. The mean data were subjected to various statistical analysis. Analysis of variance for randomized block design was carried out as per standard procedure of Panse and Sukhatme, 1978. Analysis of variance for combining ability was carried out on the parental and F₂s means following method 2 (excluding reciprocals) and model-l (Fixed effect model) of Griffing, 1956. The GCA and SCA effects for all parents and crosses respectively were calculated and significance was tested by ‘t’ test.

The results from analysis of variance (Table 1) and ‘F’ test indicated that there was existence of highly significant differences among genotypes, parents and F₂s for all the characters under study, pointing out that there was considerable genetic variability present in selected materials. The average performance of crosses was different from that of parents in characters like days to 50% flowering, plant height, number of branches, number of pods /plant, pod length, number of seeds/pod,100-seeds weight and seed yield/plant, as evident from mean square values of ‘Parents vs crosses’ for above characters. Surashe et al., 2017, Shalini and Lal, 2019, Gill et al., 2020 also reported presence of significant variability among parents and crosses. This implies that selected material was appropriate for study of combining ability and gene action involved in inheritance of different characters.
 

        
The ANOVA for combining ability of all the characters presented in Table 2 revealed that there was highly significant differences for GCA and SCA of all the characters, indicating the preponderance of both additive and non-additive genetic components of variation involved in inheritance of all these characters of yield and component traits. The crucial importance of both additive and non-additive gene effects for inheritance of characters was also supported by Barad et al., 2007, Selvam and Elangaimannan, 2010, Patil et al., 2011, Nath et al., 2018 and Samantaray et al., 2018 .
 

 
As apprehended from the ratio of 𝛔²gca and 𝛔²sca (Table 3), there was preponderance of additive gene action for the characters viz., days to 50% flowering, days to maturity and pod length and rest of the characters exhibited non-additive gene action. The characters exhibiting additive gene action were in agreement with the findings of Priya et al., 2012 and Thamodharan et al., 2017. The non-additive type of gene action for yield characters was confirmed by Selvam and Elangaimannan, 2010, Narashimulu et al., 2014 and Gill et al., 2020. Heritability estimates of all the characters were very high except for the character number of primary branches /plant, indicating that there was less influence of environmental effects and the selection will be beneficial for crop improvement.
 

 
General combining ability (GCA) effects
 
GCA effect is outcome of additive gene action and additive × additive epistatic interaction. Based on the results of GCA effects (Table 4), it was apparent that Kamdev was high general combiner for six of the yield related traits which are days to 50% flowering, days to maturity, number of primary branches/plant, number of clusters/plant, pod length and 100-seeds weight. The parent Pant Mung 8 was considered as the high general combiner for four traits which are directly related to yield. The variety OBGG 58 is the good general combiner for traits of earliness and plant height, however, showed average GCA effects for other characters. The variety MGG 385 was the good general combiner for only two characters namely pod length and number of seeds / pod and medium general combiner for seed yield. From the above, it is inferred that although no single variety was a good combiner for all the characters yet as a whole, varieties Pant Mung 8, MGG 385, Kamdev and OBGG 52 can be considered as good combiners for producing better performing crosses. Similarly, Singh and Dikshit, 2003, Barad et al., 2008, Rout et al., 2009, Selvam and Elangaimannan, 2010, Patil et al., 2011, Aher et al., 2012, Narashimulu et al., 2014, Viraj et al., 2020 and Sen et al., 2018 have isolated best general combiners on the basis of their GCA effects.
 

 
Specific combining ability (SCA) effects
 
In general, SCA does not contribute much in crop improvement of self-pollinated crops except for the crops where commercial heterosis is exploited. The SCA effects of 15 different crosses represented in Table 5 disclosed that the cross Kamdev/OBGG 58 and Pant Mung 8/OBGG 58 had desirable significant negative value for the character days to 50% flowering. For days to maturity, the desirable significant negative value was shown by crosses Kamdev/OBGG 58 and IPM-02-14/OBGG 58. For plant height, cross combinations Pant Mung 8/OBGG 58 was having negative desirable significant SCA effects. The crosses IPM-02-14/OBGG 58, Pant Mung 8/Kamdev, MGG 385/OBGG 52 showed positive significant SCA effects with high average performances for number of clusters / plant. The cross IPM-02-14/OBGG 58 was best cross for number of pods / plant. For number of seeds / pod, cross Pant Mung 8/Kamdev followed by OBGG 52/OBGG 58 and Pant Mung 8/IPM-02-14 were best combiner. The crosses Pant Mung 8/OBGG 52 followed by MGG 385/IPM-02-14 and IPM-02-14/OBGG 58 showed positive significant SCA for the character 100-seeds weight. There were five positive significant crosses for seed yield / plant and top three were Kamdev/IPM-02-14, MGG 385/OBGG 52, OBGG 52/OBGG 58. Selvam and Elangaimannan, 2010, Zuge Sopan et al., 2018 and Viraj et al., 2020 also have isolated best promising crosses on the basis of SCA effects.
 

        
Based on comparative study of per se performances and SCA values (Table 6), it specifies that not a single cross showed best SCA value for all the characters. From the table of SCA effects and per se performances it was observed that the cross, IPM-02-14/OBGG 58 was having high SCA effects and per se performances for five characters namely days to maturity, number of primary branches / plant, number of clusters / plant, number of pods / plant and 100-seeds weight, but it was significant for seed yield/plant. The next best cross was OBGG 52/OBGG 58 which was significant and having high desirable mean performance for four of the characters namely, seed yield / plant, number of seeds / pod, pod length and number of primary branches. It was followed by cross MGG 385/OBGG 52 having high SCA and mean performance for four characters like days to maturity, number of clusters / plant, pod length and seed yield / plant.
 
 
It was observed that the parental combinations in the crosses for different desirable characters like high × high, high × medium, high × low, medium × high, medium ×medium, medium × low, low × high and low × medium were in accordance with the findings of Pawale et al., 2017 and Gill et al., 2020. These desirable cross combinations involving high × high type may be due to additive type of gene actions which are fixable in nature and this type of combinations could be exploited further using simple line selection and pedigree method, which was aided by the findings of Singh and Dikshit, 2003 and Patil et al., 2011. The crosses having high SCA effects, but involving one good combiner and the other of medium or poor, might be due to epistasis like additive × dominance type of interaction which is considered as non-fixable. It was further suggested that for the characters governed by non-additive components, recurrent selection that is selection following hybridisation and inter-mating of superior parents in segregating generation will be beneficial for achieving improvement for yield and its related traits. Alternatively, it can be said that to achieve a maximum gain from selection, the selection has to be deferred to later generation of F₃ or F₄ which will reduce the distracting effect of non-additive gene action.