GENETIC VARIABILITY, CORRELATION AND PATH COEFFICIENT ANALYSIS IN GROUNDNUT (Arachis hypogaea L.)
Loading...
Date
2011-08
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
jau,junagadh
Abstract
The present investigation was carried-out to assess the genetic variability, correlation coefficients, path coefficients, selection indices and genetic divergence in 54 genotypes of groundnut (Arachis hypogaea L.). The experiment was layed-out in a randomized block design with three replications at the Main Oilseed Research Station, Junagadh Agricultural University, Junagadh during kharif-2010. The observations were recorded on 18 characters viz., days to 50% flowering, days to maturity, plant height, number of primary branches per plant, number of secondary branches per plant, number of undeveloped pods per plant, kernel yield per plant, pod yield per plant, number of pegs per plant, number of mature pods per plant, sound mature kernels, 100-kernel weight, 100-pod weight, shelling out turn, protein content, oil content, biological yield per plant and harvest index.
Analysis of variance revealed significant differences among the genotypes for all the characters studied except number of primary branches per plant. A wide range of variation was observed for most of the important yield components. High estimates of genotypic coefficient of variation was observed for number of secondary branches per plant, number of undeveloped pods per plant, number of pegs per plant, harvest index, number of mature pods per plant, kernel yield per plant, biological yield per plant and pod yield per plant. High heritability coupled with high genetic advance as per cent of mean was observed for number of pegs per plant, harvest index, number of mature pods per plant, biological yield per plant, 100- pod weight and number of secondary branches per plant, pod yield per plant, kernel yield per plant and 100-kernel weight.
The analysis of correlation coefficients suggested that the magnitude of genotypic correlations were higher than the corresponding phenotypic correlations for the most of the pairs. The pod yield per plant showed significant and positive genotypic and phenotypic correlations with kernel yield per plant, harvest index, shelling out-turn, oil content and 100-pod weight. While its association with protein content was significant but negative at genotypic level only.
The path coefficient analysis revealed the high and positive direct effects of days to maturity, kernel yield per plant, biological yield per plant and harvest index towards pod yield. The most of the characters contributed indirectly to pod yield through biological yield per plant and harvest index. Based on correlation and path analysis, harvest index, biological yield per plant, number of mature pods per plant and 100-seed weight were identified as the most important components of pod yield.
The selection indices forming 31 combinations involving pod yield and four yield components were constructed using the discriminant function technique. In a single character index, the maximum efficiency was exhibited by harvest index followed by biological yield per plant, shelling out-turn and kernel yield per plant. The efficiency of selection increased with the inclusion of more number of characters in the index. The highest relative efficiency was exhibited by a selection index involving two component characters viz., shelling out-turn and biological yield per plant (X3. X4) followed by an index based on four characters. i.e., included pod yield per plant, kernel yield per, shelling out-turn and harvest index (X1 .X2. X3. X5).
The 54 genotypes were grouped into 14 clusters by Mahalanobis’s D2-statistic. The clustering pattern of the genotypes did not confirm to the geographical distribution. The maximum inter cluster distance was found between clusters XII and XI followed by that between IX and XII and XI and X. The cluster III was superior for pod yield per plant, kernel yield per plant, shelling out-turn and harvest index, while Cluster VII was the best for days to 50% flowering, days to maturity and 100-kernel weight. Cluster V was the best for protein content. The cluster IX was good for number of secondary branches per plant, number of undeveloped pods per plant and sound mature kernels per plant. The cluster X was good for plant height and biological yield per plant. The cluster XII was good for days to mature, number of pods per plant, sound mature kernels per plant and oil content. The cluster VI was the good for 100-pod weight, while cluster XIV was good for number of primary branches per plant and number of pegs per plant. Therefore, in the present investigation, bases on high yielding genotypes and large inter cluster distances, it is advisable to attempt crossing of the genotypes from cluster XII , of cluster IX, X and XI which may lead to broad spectrum of favourable genetic variability for yield improvement in groundnut.
Overall, it can be concluded from the study of variability, correlation, path coefficient analysis, selection indices and genetic diversity that days to maturity, kernel yield per plant, shelling out-turn, biological yield per plant and harvest index were most important yield contributes and the emphasis should be given to these traits for pod yield improvement in groundnut. Selection index based on either two traits (X3.X4) or four traits (X1X2X3X5) would be useful for indirect selection.
Description
Keywords
botany