Comparative study of parametric and non-parametric methods for stability and simultaneous selection for grain yield in finger millet [Eleusine coracana (L.) Gaertn.]
Loading...
Date
2022-07
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
G.B. Pant University of Agriculture and Technology, Pantnagar, District Udham Singh Nagar, Uttarakhand. PIN - 263145
Abstract
The present investigation entitled “Comparative study of parametric and non-parametric methods for stability and simultaneous selection for grain yield in finger millet [Eleusine coracana (L.) Gaertn.]” was conducted with fifty-three germplasm accessions, that were collected from various districts of Uttarakhand, along with seven standard checks viz., VL-347, VL-149, VL-324, VL-325, VL-352, VL-315 and PRM-1. The experiment was laid in four environments, consisting of two locations, Pantnagar and Majhera over the period of two years, 2019 and 2020 during kharif season. In each trial the experiment was laid in Randomized Complete Block Design, with three replications. The data were recorded for fourteen quantitative traits. Analysis of variance (ANOVA) for each of the four environments, revealed highly significant differences among genotypes for all the fourteen quantitative traits. Likewise, at each location (Pantnagar and Majhera), the pooled analysis of variance over 2019 and 2020 seasons, also exhibited significant differences among finger millet genotypes for all the 14 traits. At Pantnagar, high heritability coupled with high GAM, exhibited in six traits viz., days to 50 % flowering, thousand grain weight, grain yield per plot, biological yield per plot, finger length and ear head length. Likewise, at Majhera, five traits viz., ear head length, finger length, grain yield per plot, finger width, biological yield per plot revealed high heritability plus high GAM, indicating prominence of additive gene effects with good predictive gains after selection for these traits. The correlation between fourteen quantitative traits evaluated in two different locations revealed that five trait pairs viz., number of fingers on main ear with grain yield per plot, number of fingers on main ear with 1000 grain weight, finger width with finger length, flag leaf blade width with plant height and days to maturity with plant height exhibited shift in inter-relationships between locations. Therefore, the inter-relationships of these traits are less reliable for selection or improvement program of finger miller at hill and plain zones. On the basis of mean performance GP2016-14 was identified as common superior genotype at Majhera and Pantnagar for grain yield per plot and biological yield per plot. Huehn’s non-parametric measures, S1, S2, S3 and S6 revealed that GP2018- 1615, GP2019-1704, GP2019-2067 and VL-352 were stable genotypes, among top fifteen ranked genotypes in each of the four parameters for grain yield per plot. Similarly, as per Thennarasu’s non-parametric stability methods i.e., N1, N2, N3 and N4, GP2018-1615, GP2019-1800, GP2019-1704, GP2019-1683 and GP2019-1912 were the common genotypes over all the four of these that placed in among the top 15 rank list for grain yield per plot. The rank-correlation values of Ecovalence (W2) and Shukla’s variance (σ2), was perfect (1.00), for all the traits, hence these showed complete congruence. Genotypes, GP2018-1649, GP2018-1619, GP2016-14, GP2019-1908 and GP2018-1613, were found stable genotypes as per superiority index, (Pi) over all the four-yield related positive traits. Rank correlation for grain yield per plot revealed significant positive association between ranks of pooled mean, superiority index, S6 and S3. Genotypes, VL- 352, GP2019-2068 and GP2019-2067 were found best for high mean and stability over all stability parameters for grain yield per plot. Based on Eberhart and Russel stability analysis over all the four traits revealed that only three genotypes namely GP2019-1912, GP2019-1918 and GP2019-1921 were found stable across all environments for all the four traits studied. Among all the parametric and non-parametric stability methods, superiority index (Pi) was found the most suitable for selection of dynamic stable
genotypes of finger millet. Next to it Huehn’s non-parametric stability estimate, S6 can also be used for the same. One of any, Thennarasu’s non-parametric stability measures, N2, N3, or N4 can be used to choose only stable genotypes that retain their yield across all environments. Eberhart and Russel model, provides the maximum information for stable and responsive genotypes with respect to the environments