**10.3 Conventional breeding methods**

*Legume Crops – Characterization and Breeding for Improved Food Security*

combining ability and their interaction with the important environmental variables. All these variables give a measure of the comparative potential of different F2 crosses. It is desirable that the progenies of only those parents be advanced beyond F2 generation that show high grain yield, yield stability, a positive general combining ability for grain yield and that are of distant genetic origin. Progenies of parents with low yield and negative general combining ability for yield must not be advanced beyond F2 generation. In an intra-species crossing programme, one parent should be a good agronomic base with higher stability and the other parent a good general combiner for yield and its components. Crosses with this strategic selection of parents are expected to give a wide range of genetic variability. To achieve stability and get a true measure of inherent genetic potential, the parental lines must be tested over a number of locations and get their combining ability

Fifty years of conventional approach of engineering different yield components in mungbean to build up a new plant type with higher productivity levels has thus far given only modest yield gains over the traditional cultivars. This approach has failed to break the present yield barriers as a whole and bring changes of scale. Based upon correlation analysis of various yield components, selections have been based mainly on number of pods per plant, seeds per pod and 100-seed weight and sometimes also on number of pods per bunch and branches per plant or podding per unit area usually called 'Pod Index'. Pods per plant is by far the most important yield component and almost all the workers have found it having positive correlation with seed yield. It is the best selection index for seed yield and could be increased by increasing number of branches per plant or number of bunches per plant or by increasing the number of pods per bunch. Most of the work has shown that branch number per plant is negatively correlated with seed yield, but bunches per plant has mostly been found to be positively correlated with seed yield. Ramanujam [75] and others have found that pods per bunch and bunches per plant both are positively correlated with yield. Increasing pods per bunch is physiologically constrained in grain legumes owing to fall of flowers and unripe or partially filled pods. It seems the most feasible path to increase seed yield is through increasing number of bunches per plant. This in essence means a plant with more number of nodes with a shorter internode length, with three to four erect branches emerging from the lower to lower-middle nodes at around 20–30° angle with the main stem, and sympodial bearing of pod inflorescences coming from the upper nodes of the main stern, each carrying around 8–10 pods. Number of seeds per pod has been shown mostly positively correlated with seed yield but many workers show it to be negatively correlated with yield. However, an optimal level of 14–16 seeds per pod should be a breeders objective. Seed weight is generally negatively correlated with seed yield but some results have shown it to be positively correlated. The strength of the newly developed second generation varieties like Pusa Vishal and Pusa Ratna lies in the fact that they have more seeds per pod (12–16) with higher 100-seed weight (5.0–5.5 g) without compromising on the pod number per plant. Many researchers advocated cereal mimics with sympodial bearing and suggested increasing pods per plant through the path of increasing the average number of pods per node and building up a soybean like plant type in mungbean [76]. He found the main stem bearing under the control of a single recessive gene and normal conventional bearing to the incompletely dominant. Plant height has been found positively

**58**

estimates.

**10.2 Component breeding**

Most of the high yielding varieties of mungbean bred and released so far have been developed through single cross pedigree method of selection. The single plant selections made in the early generations restrict carrying forward the bulk of created variability, which gets lost quickly giving way to homozygosity with each succeeding generation. This method has served the mungbean improvement programme well in the past, but lately no productivity advances are materializing due to the inherent genetic limitation of the method. The intermitting of selected F2 plants and selections in the late generations will help to harness most of the desirable genes.
