ICAR-AICRP on Pearl Millet, Agricultural Research Station, Swami Keshwanand Rajasthan Agricultural University, Bikaner – 304006, Rajasthan, India Corresponding Author: Aman
Also available as PDF (figures and tables appear in the PDF).
Pearl millet (Pennisetum glaucum L.) is a vital cereal in arid and semi-arid regions due to its tolerance to high temperature, moisture stress and poor soil fertility as well as its nutritional and fodder value. A study was conducted at ARS, Bikaner under AICRP on Pearl Millet to evaluate 23 pearl millet entries including five standard checks in the harsh agro-climatic conditions of western Rajasthan to identify early-maturing and high-yielding genotypes. The experiment, laid out in a randomized block design with three replications, recorded ten quantitative traits along with disease incidence. Significant variation was observed among genotypes for all traits indicating substantial genetic variability. High phenotypic and genotypic coefficients of variation were recorded for grain yield while moderate values were observed for dry fodder yield and panicle diameter. Most traits exhibited high heritability with grain yield and dry fodder yield showing both high heritability and genetic advance, suggesting predominance of additive gene action. Plant height, panicle length, panicle diameter and 1000-seed weight showed high heritability with moderate genetic advance. No incidence of major diseases was recorded, indicating overall genetic stability. The findings reveal considerable potential for selection and improvement of key traits and demonstrate the effectiveness of phenotypic selection in developing superior pearl millet genotypes adapted to arid environments.
Keywords: Pearl millet; genetic variability; heritability; genetic advance; grain yield; dry fodder yield; phenotypic selection; Pearlmillet; geneticvariability; geneticadvance; grainyield; dryfodderyield; arid regions; phenotypicselection
Pearlmilletstandsasavitalcoarsecerealinthearid andsemi-aridtropics. Itisvaluednotonlyforitsresilienceunderconditionsofhightemperature,lowrainfall and poor soil fertility but also for its nutritional profilethatisrichinenergy,protein,dietaryfiberand mineralssuchasironandzinc(Gupta etal.,2015;Hassan et al., 2021; Karthik et al., 2024). Furthermore, itsversatilitymakesitsuitableforbothfoodandfodder.WithinIndia,pearlmilletholdsaprominentpositionparticularlyinRajasthan,Gujarat,Haryana,Maharashtra and parts of Uttar Pradesh (Satyavathi et al.,2021).However,WesternRajasthanfacesextreme agriculturalconstraintsincludingvariablerainfall,high evapo-transpiration,sandysoilsandfrequentheatand moisturestress(Singh,2025).Despitethesechallenges, itremainsakeycropintheregionwhichisessential tolocalfarmingsystems.Consequently,limitedirrigationandpoorsoilhealthresultinproductivitylevels thatfallwellbelowstateandnationalaverages(Singh, 2025;YadavandRai,2013).ZoneIc(Bikaner,Jaisalmer andChurudistricts)reported4.23lakhhectaresofarea, 2.44lakhtonnesofproductionandaproductivityof 537kg/ha;incontrast,Rajasthanasawholereported 3.52lakhhectaresofarea, 53.73lakhtonnesofproductionand1234kg/haproductivityduringthesame period(Anonymous,2025).Thisstarkproductivitygap underscoresthenecessityforlocation-specificgenotypeselection(YadavandRai,2013),asagro-climatic gradientsrevealthatwesterndistrictsconsistentlylag behindtherestofthestate.
AllIndiaCoordinatedResearchProjectonPearlMillet focusesonidentifyinghigh-yielding,stableandadapted genotypesforvariousagro-climaticzonestoaddress theseissues(Satyavathi etal.,2021). ZoneA1,representingtherainfedregionsofRajasthan,Haryanaand Gujarat,exhibitsnotablylowproductionandproductivity.Beyonditsgrainvalue,thecropprovidesessentialfodderforlivestockandsupportsdietarydiversity amongresource-poorpopulations(Karthik etal.,2024). Economically,itservesasacriticallivelihoodsourcefor smallholders(Naorem etal.,2023).Thepresentinvestigationwasconductedtoidentifypromisingearlypearl milletentriessuitedtotheharshandaridenvironment ofZoneA1bycomparingthemwithstandardchecks. Suchevaluationisvitalbecausegenotypesthatthrive inoneenvironmentmayfalterundertheextrememoistureandheatstresscharacteristicofwesternRajasthan. Therefore,testinginthetargetenvironmentisessential toselectgenotypeswithsuperiorearliness,stability andyield.
Theexperimentalmaterialcomprising23pearlmillet entries(IHT101toIHT123), includingfivestandard checks,wasevaluatedattheAgriculturalResearchStation(ARS),SwamiKeshwanandRajasthanAgricultural University(SKRAU),Bikaner,Rajasthan.Thetrialwas conductedinaRandomizedBlockDesign(RBD)with threereplications.Eachentrywasgrowninanetplot measuring4.0m × 1.80m,maintainingacropgeometryof60cm × 15cm. Theexperimentalsiteischaracterizedbysandysoilwithanalkalinereaction(pH 8.0).Theavailablenutrientstatusofthesoilwasvery −1 lowinnitrogen(80kgha ),lowinphosphorus(18kg −1 −1 ha )andmediuminpotassium(168kgha ). Recommendedagronomicpracticesfortheregionwere followedthroughoutthecropseason.Basaldosesof nitrogenandphosphoruswereappliedatsowingwhile theremainingnitrogenwassuppliedthroughtopdressingattheappropriatecropgrowthstage.
Observationswererecordedontenquantitativecharacters, namely: days to 50% flowering, days to maturity,plantheight(cm),paniclelength(cm),panicle diameter(cm),1000-seedweight(g),populationatharvest(numberpernetplot),seedsetunderbagging(%), grainyield(kgpernetplot)anddryfodderyield(kg pernetplot).Grainandfodderyieldsrecordedonanet plotbasiswereconvertedtohectareyieldusingaconversionfactorof1388.89.Inaddition,theincidenceof majorpearlmilletdiseases, viz.,downymildew,smut andergot,wasrecordedundernaturalfieldconditions andexpressedaspercentageinfection. Theidentity detailsofallentriesarepresentedinTable1.Genetic variabilityparameterswerecomputedusingstandard biometricalprocedurestoevaluatethemagnitudeof variationavailableforselectionandgeneticimprovementinpearlmillet.
Therecordeddataweresubjectedtostatisticalanalysis toestimatemeanperformanceandassesstheextent ofgeneticvariabilityamongthehybridentries.OPSTAT statisticalsoftwarewasusedtoanalyzetherecorded dataat1%and5%levelofsignificance.
The variability among genotypes was estimated usingrange,mean,standarderror,phenotypicvariance, genotypicvarianceandcoefficientsofvariation.The variancecomponentswerecomputedfollowingstandardproceduresandusedtoestimatephenotypicand genotypiccoefficientsofvariation,heritabilityandgeneticadvance.
Genotypicvariance(σg )wascalculatedas:
PusaHybrid IHT101 MH2887 411A × PPMI1327 ICAR-IARI,NewDelhi Bajra2501 IHT102 MH2888 DHBH-24305 DHLB-14A × DHLBI-1035
IHT103 Check PB1756 PSP91xPP107 ProtectionLtd.,
IHT104 MH2889 DHBH-24308 DHLB-36A × DHLBI-1035 ICAR-ICAR-AICRP-PM,Dhule IHT105 MH2890 RHB284 ICMA02333 × RIB494 ICAR-AICRP-PM,SKNAU,Jaipur IHT106 MH2891 RHB285 ICMA04888 × RIB20K86 ICAR-AICRP-PM,SKNAU,Jaipur IHT107 MH2892 RHB287 ICMA99444 × RIB494 ICAR-AICRP-PM,SKNAU,Jaipur IHT108 MH2893 RHB288 ICMA93333 × RIB3135-18 ICAR-AICRP-PM,SKNAU,Jaipur IHT109 MH2894 RHB290 ICMA98222 × RIB15177 ICAR-AICRP-PM,SKNAU,Jaipur IHT110 Check RHB223 ICMA96666xRIB3135-18 ICAR-AICRP-PM,SKNAU,Jaipur IHT111 MH2895 RHB292 ICMA04999 × RIB15S076 ICAR-AICRP-PM,SKNAU,Jaipur IHT112 MH2896 BHB2501 ICMA88004 × BIB-62 ICAR-AICRP-PM,SKRAU,Bikaner IHT113 MH2897 BHB2502 ICMA98222 × BIB23-1 ICAR-AICRP-PM,SKRAU,Bikaner IHT114 MH2898 MPMH52 ICMA95222 × MIR519-1 ICAR-AICRP-PM,PCUnit,Jodhpur IHT115 MH2899 MPMH53 ICMA04999 × MIR519-1 ICAR-AICRP-PM,PCUnit,Jodhpur IHT116 Check HHB67(Imp.) ICMA843-22xH77/833-2-202 ICAR-AICRP-PM,CCSHAU,Hisar IHT117 MH2900 PA9070 PSP140 × PP125 CrystalCropPrtectionLtd. IHT118 MH2901 CZH278 ICMA04999 × CZI2022/7 ICAR-CAZRI,Jodhpur IHT119 MH2902 CZH279 ICMA843-22 × CZI2023/4 ICAR-CAZRI,Jodhpur IHT120 Check MPMH35 ICMA194555xMIR1252 ICAR-AICRP-PM,PCUnit,Jodhpur IHT121 MH2903 CZH280 ICMA083333 × CZI2023/4 ICAR-CAZRI,Jodhpur IHT122 Check 86M94 9PHGF39AxM444R PioneerOverseasCorp.,Hyderabad IHT123 MH2904 RVBH-2253 ICMA97111 × 24110 ICAR-AICRP-PM,RVSKVV,Gwalior
σg = (MST − MSE)/r where,MST=meansquareduetotreatments(genotypes),MSE=errormeansquare,and r =numberof replications.
Phenotypicvariance(σ )wasestimatedas: p 2 2 2 σp = σg + σe where, σe =environmental(error)variance.
AccordingtoSinghandChaudhary(1999),thephenotypiccoefficientofvariation(PCV)andgenotypiccoefficientofvariation(GCV)werecalculatedusingthe followingformulas: √︂ PCV(%) = ( σp/X) × 100 √︂ GCV(%) = ( σg/X) × 100 where, isthegrandmeanofthetrait.
X ¯
ThevaluesofGCVandPCVwereclassifiedaslow(<10%), moderate(10–20%),andhigh(>20%)followingthecriteriasuggestedbyDeshmukh etal. (1986).
Heritabilityinthebroadsense(h )wasestimatedfor eachtraitaccordingtoAllard(1960)as:
ICAR-ICAR-AICRP- PM,Dhule CrystalCrop
Hyderabad 2 2 2 h (%) = (σg /σp) × 100 Theexpectedgeneticadvance(GA)underselectionwas calculatedfollowingJohnson etal. (1955)assuminga selectionintensityof5%: √︂ 2 2 GA = k × h × σ p where, k =standardizedselectiondifferential(2.06at 5%selectionintensity), h =heritabilityinbroadsense, √︂ and σp =phenotypicstandarddeviation.
Geneticadvanceaspercentofmean(GAM)wascomputedas: GAM(%) = (GA ¯) 100 /X × 2 ¯ where,GA=geneticadvanceand =grandmeanof X ¯ 2 ¯ thetrait.
TheANOVAresultsarepresentedinTable2.Themean sumofsquaresduetotreatmentsaresignificantforall thecharactersstudied,whichrevealedsignificantdifferencesamongthegenotypesforallthetraitsstudied. Significantgeneticvariabilityinbreedingmaterialisa prerequisiteasitnotonlyprovidesthebasisforselectionbutalsooffersvaluableinformationforidentifying diverseparentsforuseinhybridizationprogrammes (Gupta etal.,2015;Singh etal.,2016;Basavaraj etal., 2017). ThemeanperformanceisgiveninTable3for eachcharacterstudied.
Allthetraitsstudiedexhibitedawiderangeofvariation,indicatingsubstantialgeneticdiversityandample scopefortheirimprovementthroughselection. 1.Daysto50%flowering: Themeandaysto50%floweringwas45.12dayswitharangeof42.33to50.00days. TheearliestfloweringgenotypewasHHB67(Imp.)(check) (42.33days)followedbyCZH278(43.00days)whereas RHB 223 (check) was the latest flowering genotype (50.00days).Basedonthecriticaldifference(CD=1.77 days),tenentriesareatparwiththeearliestflowering entry.
2. Days to maturity: The overall mean for days to maturitywas74.99daysandvariedfrom72.00to79.00 days. HHB67(Imp.) (check)maturedearliest(72.00 days)whereasRHB223(check)maturedlatest(79.00 days).Thedifferencebetweentheearliestandlatest maturinggenotypesexceededtheCDvalue(2.18days), indicatingsignificantvariationformaturityduration. 3.Plantheight: Plantheightrangedfrom116.00cm to 178.33cm with ameanof 158.87cm. Thetallest genotypewasRHB292(178.33cm)followedbyMPMH 52(177.67cm),whileRHB223(check)wastheshortest genotype(116.00cm).Basedonthecriticaldifference (CD=13.15cm),sixentriesareatparwiththetallest entry.
4.Paniclelength: Themeanpaniclelengthwas23.48 cm,rangingfrom20.67to26.67cm.ThelongestpanicleswererecordedinMPMH52(26.67cm)whereas DHBH-24305(IHT102)exhibitedtheshortestpanicles (20.67cm).Basedonthecriticaldifference(CD=2.30 cm),sevenentriesareatparwiththelongestpanicle. 5. Panicle diameter: Panicle diameter varied from 2.10to3.07cmwithameanof2.55cm.Themaximum paniclediameterwasobservedinBHB2501(3.07cm), followedbyPA9070(3.03cm),whileRHB287recorded theminimumdiameter(2.10cm).Basedonthecritical difference(CD=0.32cm),threeentriesareatparwith themaximumpaniclediameter.
6.1000-seedweight: Theaverage1000-seedweight was9.78gandrangedfrom7.87to11.33g.Thehighest testweightwasrecordedinPB1756(check)(11.33g), followedbyPusaHybridBajra2501(IHT101)(11.03g), whereasMPMH52andCZH280exhibitedthelowest value(7.87g).Basedonthecriticaldifference(CD=1.01 g),sixentriesareatparwiththehighesttestweight. 7.Populationatharvest: Themeanplantpopulation atharvestwas72.38plantspernetplot,rangingfrom 67.00to77.33plants.ThehighestpopulationwasobservedinRHB223(IHT110),BHB2501(IHT112),and RHB287(IHT107)(77.33plants), whilePusaHybrid Bajra2501(IHT101)recordedthelowestpopulation (67.00plants).
8.Seedsetunderbagging: Themeanseedsetunder baggingwas50.94%,withvaluesrangingfrom43.33% to58.33%.ThehighestseedsetwasobservedinDHBH- 24305(IHT102)andCZH279(IHT119)(58.33%),while PB1756(IHT103)recordedthelowestseedset(43.33%). 9.Grainyield: Grainyieldpernetplotaveraged0.55kg andrangedfrom0.35to0.83kg.Thehighestgrainyield wasrecordedbyPusaHybridBajra2501(0.83kg),followedbyBHB2501,MPMH53and86M94(check)(0.75 kgeach)andRHB292(0.72kg).Basedonthecritical −1 difference(CD=0.15kgplot ),thegrainyieldsofBHB 2501,MPMH53,86M94andRHB292werestatistically atparwiththehighestyieldinggenotype.Thelowest grainyieldwasobservedinRHB223(check)(0.35kg). 10. Dryfodderyield: Dryfodderyieldrangedfrom 2.12to3.57kgpernetplot,withameanof2.54kg.The highestdryfodderyieldwasrecordedinMPMH52(3.57 kg),followedbyRHB223(check)andMPMH53(3.27kg each).RHB223andMPMH53werestatisticallyatpar withMPMH52.Thelowestfodderyieldwasobserved inPusaHybridBajra2501(2.12kg).
Noincidenceofdownymildew,smutorergotwasobservedinanyofthetestedentriesthroughoutthecrop growth period under natural field conditions. Thus, allgenotypesexhibitedfreedomfromthesemajordiseasesduringthecourseofexperimentation.
Table4depictstheestimatesofgenotypiccoefficient ofvariation(GCV),phenotypiccoefficientofvariation (PCV),heritability(broadsense),geneticadvance(GA) and genetic advance as per cent of mean (GAM) for differentcharactersstudiedinpearlmillet.
Estimatesofphenotypiccoefficientsofvariationwere higherthangenotypiccoefficientofvariation,whichdepictstheenvironmentalinfluenceoncharacters.The highestimatesofPCVandGCVwerefoundforgrain yieldas26.71and21.18,respectively.Thesefindings areconsistentwiththeresultsreportedbyChaudhary etal. (2012),Gupta etal. (2015)andRamya etal. (2022). Moderate estimates of PCV and GCV were recorded for dry fodder (17.30 and 14.00) and panicle diameter(12.65and10.05).Certaincharactersshowedmoderate estimates for PCV such as seed set under bagging(12.99),1000-seedweight(11.50)andplantheight (10.23). ThesefindingsareinagreementwiththereportsofSingh etal. (2014a),Kumar etal. (2015),Sharma etal. (2018)andRamya etal. (2022). Althoughthese characters,alongwiththeremainingcharactersstudied,depictedlowestimatesforGCV.
Heritabilityisanimportantgeneticparameterthatmeasurestheproportionofphenotypicvariationattributable togeneticfactorsandindicatesthelikelihoodoftransmittingtraitsfromparentstotheiroffspring.Basedon heritabilityestimates,traitsarecategorizedashaving low(<30%),moderate(30–60%)orhigh(>60%)heritability(Robinson etal.,1949).Estimationofheritabilityassistsplantbreedersinidentifyingandselecting superiorgenotypesfromgeneticallydiversepopulations. Traitsexhibitinghighheritabilityaregenerally lessinfluencedbyenvironmentalfactorsandcanbe improvedeffectivelythroughselection.Inthepresent study,mostofthetraitsexhibitedhigh(62.89%–78.27%) heritabilityestimates.Theseresultsareinagreement
seedsetunderbagging,indicatingagreaterinfluence ofenvironmentalfactorsontheexpressionofthese traits. SimilarfindingswerereportedbyRama etal. (2005)andKumar etal. (2016),asheritabilityisagood index of transmission of characters from parents to theirprogeny.
Geneticadvanceisanimportantparameterthatindicatestheexpectedimprovementinatraitachieved throughselection.Inthepresentstudy,highheritabilityandgeneticadvanceaspercentageofmean(>20) wasrecordedfordryfodderyieldandgrainyield,indicatingpredominanceofadditivegeneactionforthese characters.Traitsdisplayingbothhighheritabilityand highgeneticadvanceasapercentageofthemeanare
Daysto50%flowering 4.509 Daystomaturity 2.565 PlantHeight(cm.) 8.912 PanicleLength(cm.) 7.812 PanicleDiameter(cm.) 10.053 1000-Seedwt.(g) 9.639
primarilyunderadditivegenecontrol,renderingthem highlyreliablefordirectphenotypicselection(Sangwan etal.,2019).1000-seedweight,paniclediameter, panicle length and plant height exhibited high heritabilitycoupledwithmoderategeneticadvanceaspercentageofmean.SimilarresultswerereportedbyLakshmana etal. (2003),Sumathi etal. (2010),Vinodhana etal. (2013),Kumar etal. (2014)andBasavaraj etal. (2017).Traitsexhibitinghighheritabilityalongsidelow geneticadvanceareofteninfluencedbyenvironmental fluctuations, necessitating more nuanced breeding approaches like heterosis or recurrent selection toachievefurthergains(Ahmad etal.,2018).Furthermore,thepresenceofhighheritabilitycoupledwithsignificantgeneticadvanceobservedforkeyyield-related traitsindicatesthatphenotypicselectionwillbeparticularlyefficientfordevelopingsuperiorgenotypes (Ayesha etal.,2019;Subbulakshmi etal.,2022).
Thesuccessofcropbreedingprogrammeslargelydependsontheavailabilityofgeneticvariabilityandthe inheritance of desirable traits, while the analysis of geneticvariationhelpsbreedersformulateeffectiveselectionstrategiesfortraitimprovement.Inthepresent study,thetraitsexhibitedvaryinglevelsofvariability, heritabilityandgeneticadvanceasapercentageofthe meanamongthestudiedgenotypes. Traitsshowing highvaluesofphenotypiccoefficientofvariation(PCV) andgenotypiccoefficientofvariation(GCV)indicatea highlevelofgeneticvariability. Highheritabilityvaluesindicatethatthetraitsunderstudyarelargelycontrolledbygeneticfactorsandarelessaffectedbyenvironmentalinfluencesintheirexpression.Thepresence ofhighvariabilitysuggeststhepotentialforeffective selectionandimprovementoftraits.Traitsexhibiting highheritabilityalongwithhighgeneticadvanceasa percentageofthemeanindicatethepredominanceof additivegeneaction,suggestingthatthesetraitscan beeffectivelyimprovedthroughselection.Incontrast, traitswithmoderateheritabilityandlowgeneticad-
5.097 78.271 3.707 8.218 3.109 68.061 3.269 4.359 10.226 75.965 25.422 16.002 9.803 63.492 3.01 12.822 12.647 63.194 0.419 16.463 11.496 70.302 1.628 16.648 7.091 10.423 1.102 1.522 12.993 6.053 0.825 1.62 17.3 65.458 0.592 23.327 26.708 62.892 0.192 34.602
vanceasapercentageofthemeanarepredominantly governedbynon-additivegeneaction,implyingthat directselectionforthesetraitsmaybelesseffective.