ABSTRACTThe present study was carried out to probethe direct and indirect selection indices in fifteen chickpea (Cicer arietinumL.
) genotypes. The experiment was conductedat Gram Breeding Research Station, Kallurkot, Pakistan, during Rabi season of2016-17. Basic analysis of variance revealed that all the genotypes were significantlydifferent (p?0.01) for all the traits.
The path coefficient analysis showedthat the harvest index had greatest direct effect on yield (0.527) which was followedby number of pods plant-1 (0.498) and 100 seed weight (0.452).Correlation coefficient studies showed highly significant correlation betweenharvest index and yield (0.
941) followed by number of pods plant-1(0.924) and 100 seed weight (0.502) signifying the importance of these traitsfor crop improvement. Correlation and path coefficient analysis studies showedthat harvest index, number of pods plant-1 and 100 seed weight havepositive and significant correlation to grain yield coupled with high directpositive effect. Therefore, these traits may be focused while attempting chickpeagenetic improvement program.Keywords: Correlation,chickpea, direct, indirect indices,grain yield.1. INTRODUCTIONChickpea is one of the first grain legume crops to be domesticated byhumans in old world (Van der Maesen, 1972).
It is cheap and rich source of protein (20%), along with 60% carbohydratesand 1.6% fats (Ali and Ahsan, 2012). The crop is being cultivated in more than30 countries of almost all the continents including Asia, Europe, North America,South America, Australia and Africa (Mushtaq et al., 2013). Chickpea is the importantrabi pulse crop contributing about 76% share in total pulses production ofPakistan, occupying > 5 % of the area under rabi crops. But the crop hasbeen facing from 15-18% decline in production due to erratic trends in amountand frequency of rainfalls, as the crop’s major share is obtained from Thalregion solely dependent on rain (PakistanEconomic Survey 2015-16). Irrespective of the fact that the crop rankssecond (after India) in terms of acreage, productivity of the crop on per unitarea basis in Pakistan is far below (276 kg ha-1) than the average world’s production of 952 kg ha-1(FAO STAT, 2014).
Therefore, the evolution of potent cultivars of chickpea is direneeded to uplift the yield on existing area.Determination of type and strength of association among variousyield traits and their impact on the final yield is imperative for selection ofsuitable germplasm for fruitful breeding programs. For improvement in chickpeagenotypes it is necessary to understand the genetic basis of yield and relatedtraits. The path coefficient analysis for yield contributing traits will behelpful in sorting out the most effective plant variable on which the selectionshould rely upon.In most of the biological systemsvarious traits are interlinked with each other to produce the final phenotype.The complexity increases with increasing number of factors interlinked in thesystem.
So identification of the most related and effective parameter forselection becomes a difficult task for a scientist attempting to improve thatparticular system. Under such complexity, the path coefficient analysis helpsto measure the direct and indirect contributions of linked traits towards theultimate yield (Singh et al., 1990). Path coefficient analysis quantifies theimpact of various traits directly and indirectly on grain yield (Gull, 1995;Bakhsh et al., 1998).Correlation analysis is very helpful indeciding the most effective selection criteria for improving yield and its contributingtraits (Khan & Qureshi, 2001). Many researchers such as Noor et al.
(2003); Farshadfar &Farshadfar (2008); Atta et al.(2008); Sharma & Saini (2010) and Ali et al. (2011) emphasized the utility of correlation analysis.Islam et al.
(1984) documented that grain yield is highly and positivelycorrelated with pods per plant, biomass and secondary branches per plant and gaveemphasis on these parameters as to serve as selection criteria in chickpea. Thecorrelation among yield and yield attributes has been widely studied (Lokendraet al. (1999). Saleem et al., 1999; Saleem et al. 2002). and Yucel et al.(2006) who reported a significant and positive correlation between pods plant-1and harvest index with grain yield plant-1.
Information acquiredfrom path coefficient analysis and correlation is very useful for a plant breederto devise an efficient selection criterion in breeding program aiming at increasedyield. The main focus of the present investigationwas to analyze the mutual relationship in various metric traits and theirdirect and indirect influence on the final economic yield (grain yield) throughpath coefficient and correlation analysis which could lead as a directionalmodel for determination of an effective selection criteria to evolve the mostefficient and potent genotypes.2. MATERIALSAND METHODSThe study for the investigation ofselection criteria in chickpea genotypes using path analysis and correlationwas carried out at Gram Breeding Research Station, Kallurkot, Pakistan, during Rabiseason 2016-17.
The experimental material comprised of fifteen chickpea genotypesincluding 4 commercial varieties Bhakkar-11, CM-2008, Noor-13 and Bittle-16 and12 elite strains viz; CH60/10, CH73/10, CH86/10, K010-10, K044-11, K065-11,CC9899, CH 85/06, DO 80-10, DO 72-11 and DO 88-11). The layout was done in randomizedcomplete block design (RCBD) with three replications. All experimental plotscomprised of 30 centimeter apart 4 rows of 4 meter length.
The seed ofgenotypes was sown with the help of dibbler making 10cm apart holes. Two seedsper hole were sown and later thinned to 1 plant per hole after 10 days ofgermination. Insecticide Emamectin @ 200 ml acre-1 was sprayed twicewith an interval of 20 days against pod borer attack during the pod formationstage. Datawas recorded for number of days taken to flowering (50% ), number of days taken to maturity (90%), plantheight (cm), number of pods plant-1,100 seed weight (g) and grain yield (Kg ha-1). Harvest index was calculatedas economic yield over total biomass.
Analysis of variance was done following Steel et al. (1997) forthe estimation of genetic differences among genotypes for the concerned traits.Path coefficients analysis was done following Dewey and Lu (1959). While, the genotypicand phenotypic correlation coefficients were calculated following Singh andChaudhry (1979).