Genetic variability and inheritance studies for ploHw tolerance in tropical and sub-tropical maize gearmsmpl By Kesbell Kaswela Eston Kaonga A thesis submitted in accordance with the requirnetms efor the degree Philosophiae Doctor in the Department of Plant Sciences, Divni sPiolant Breeding, in the Faculty of Natural and Agricultural Sciences at the Univer soift ythe Free State Bloemfontein, South Africa 2015 Promotor: Prof. Maryke T. Labuschagne (PhD) Co- Promotors: Dr. Amsal Tarekegne (PhD) Dr. Angeline van Biljon (PhD) DECLARATION I, Kesbell K.E. Kaonga, do hereby declare thatt hthees is hereby submitted for qualification for the degree Philosophiae Doctor in Agriculturte t hae University of the Free State represents my own original, independent work anadt It have not previously submitted the same work for a qualification at another univer/s fiatyculty. I further more cede copyright of the thesis in fuarv of the University of the Free State. ............................................... . . . . .......................................... Kesbell Kaswela Eston Kaonga Date ii DEDICATION This work is specifically dedicated to our last nb odraughter Deliness Kaonga, who misssed parental care during the course of my study, selcyo tnod my wife Judith Lwesha Kaonga, our son Arisai Kaonga, and our daughters Byenadla C alenvereen Kaonga for their patience and hard times they may have gone through durin gs tmudyy period. iii ACKNOWLEDGEMENTS To Almighty God, the Creater and the one who tackaeres of my life, thank you for keeping me healthy throughout my study period. It is becea oufs You that I have been able to complete my studies. I wish to sincerely thank Mthien istry of Agriculture, through the former Principal Secretary, Dr. Andrew Daudi forf eorifng me a Government PhD Scholarship. I don’t take this for granted knowitnhga t I was among the very few beneficiaries. I would like to thank the ministroyr fthe financial support and adminstrative clearance to enable me to undertake the study. I am indebted to the University of the Free StaDtep, artment of Plant Sciences: Plant Breeding, for accepting and registering me as tshteuidr ent. Again I don’t take this for granted because I was among the few students tehraet ewnrolled during the period. My gratitude goes to the Department of Agriculture eRaercsh Services (DARS) for adminstrative clearance and moral support and erangcoeument during the entire period of my study. Thank you Dr. A.P. Mtukuso, Dr. Banda aHnudman Resource staff for administrative issues during my study period. I wish to convey my sincere gratitude and appreiocnia to various organisations and individuals who contributed in one way or anothne tre irms of resources and knowledge. It is not possibe to mention the names of all indiavilds uand institutions but your valuable contributions have been fully recognised and apiaptredc:  CIMMYT-Colombia through Dr. Luis Narro for the maei zgenotypes used in the study. Populations, released and non-releasedd i nlibnrees.  CIMMYT-Zimbawe through Dr. Amsal Tarekegne for tmhea ize genotypes used in the study. Populations, released and non-redle iansbered lines.  IITA - Nigeria through Dr. Abebe and Dr. Apraku ftohre maize populations as well a the detailed description of maize genotywpheisc h originated from their instution through CIMMYT-Zimbabwe.  The Soils and Agricuture Engineering Research Codmitmy oTeam through Dr. W. Makumba and M. Munthali for guidance in the hoypdornic nutrient solution experiment. Laboratory technicians for the nutr iseonltution preparations and field soil sampling and laboratory analyses. iv  The Maize Research Commodity Team Technical StMafaf iz(e Breeding and Agronomy) for the setting up of the hydroponic nieuntrt solution experiment in a glasshouse transplanting and initial data collenc atinod final data collection.  Lilongwe Water Board for the support in distilleda twer when demand was high to be met by Chitedze Soils and Agricuture Engiinege Lrab.  Maize technicians, research attendants and stamtiaona gers for all research stations that hosted the field trials: Lunyangwas eRaerch, Meru Research, Baka Research, Chitedze Research, Bembeke Research, bBwveu mResearch, Tsangano Research site and Chitala Research.  Prof M.T. Labuschagne for her excellent supervi sainodn encouragement, material and other valuable support.  Dr. Amsal Tarekegne of CIMMYT-Zimbabwe for guidan icne the breeding work in Malawi and for supervision.  Dr. Angeline van Biljon (PhD) for supervisiond a snupport rendered on recent publications on research done on stress tolera nce.  Dr. B.M. Jumbo of CIMMYT-Kenya for accepting ando swhing interest to edit papers earmarked for publication from this work.  Me. S. Geldenhuys of the Plant Breeding office ,a flol rthe communications, other adminstrative issues, moral support and encourangte dmuering the period of my study.  My fellow PhD students in the Plant Science Depeanrttm for their cooperation and assistance, academically and socially. v CONTENTS DECLARATION ii DEDICATION iii ACKNOWLEDGEMENTS iv CONTENTS vi LIST OF TABLES xi LIST OF FIGURES xiv LIST OF APPENDICES xv ABBREVIATION AND SYMBOLS Xvi CHAPTER 1 1 General introduction 1 1.1 Origin, importance and production constrainf tms aoize 1 1.2 Maize production in Malawi 2 1.3 Maize agro-ecology in Malawi 2 1.4 Abiotic constraints to maize production in Mwail a 5 1.5 Biotic constraints to maize production in Mail aw 7 1.6 Malawi National Maize Breeding Programme 8 1.7 References 9 CHAPTER 2 14 Literature review 14 2.1 Importance of maize and consumption levels 14 2.1.1 Important abiotic factors affecting maized purcotion 14 2.2 Concept of low pH, definition and origin 15 2.2.1 Research findings on aluminium toxicity etfsfe c 16 2.3 Mechanisms for low pH tolerance 17 2.3.1 Genes and inheritance for tolerance to aliumi ntoxicity 18 2.3.2. Genetic variability in various crops for maliunium tolerance 19 2.4 Types of mechanisms for aluminium tolerance 21 2.4.1 Physiological mechanisms of aluminium tolecrea n 22 2.4.2 Genetic mechanism for aluminium tolerance 23 2.5 Use of modern tools in breeding for low pH rtaonlece: QTLs, marker 24 assisted selection and transgenic’s 2.6 Diallel evaluation 25 2.7 Combining ability analysis 26 2.7.1 General combining ability analysis 26 2.7.2 Specific combining ability analysis 26 2.7.3 Importance of combining ability analysis 27 2.7.4 Research findings on combining ability stus dinie maize 27 2.8 Heritability estimation 28 2.8.1 Importance of narrow sense heritability 29 2.8.2 Research findings on heritability studiesm iani ze 29 2.9 Heterosis 30 2.9.1 Research findings on heterosis studies inz em ai 31 vi 2.10 Correlations 31 2.10.1 Research findings on correlation in maize 32 2.11 Stability analysis 33 2.11.1 Stability definition and its concept 33 2.11.2 Phenotypic stability analysis techniques 34 2.11.2.1 Cultivar performance technique for estimating cvualrt istability 35 2.11.2.2 Wricke’s ecovalence technique for estimating cualrt isvtability 35 2.11.2.3 Shukla’s stability variance parameter for estimga tcinultivar stability 35 2.11.2.4 Regression coefficient and deviation mean squares 6 3 2.11.3 Multivariate techniques for stability anaisly s 37 2.11.3.1 Additive main effects and multiplicative interacnti oanalysis technique 37 2.12 References 39 CHAPTER 3 51 A hydroponic nutrient solution experiment for tensgt ilow pH tolerance in tropical and 51 sub-tropical maize genotypes 3.1 Abstract 51 3.2 Introduction 51 3.2.1 Hydroponic nutrient solution 52 3.2.2 Justification for use of hydroponic nutriesnotlu tion experiment 53 3.3 Materials and methods 54 3.3.1 Experimental materials 54 3.3.2 Experimental procedure and design 57 3.3.3 Nutrient solution preparation 58 3.3.4 Data collection, measurements and calcul aotfi odnerived data 58 3.3.5 Statistical analysis 59 3.3.5.1 Analysis of variance 59 3.4 Results 59 3.4.1 Observed symptoms of aluminium toxicity 59 3.4.2 Analysis of variance 60 3.5 Discussion 65 3.6 Conclusions and recommendations 66 3.7 References 67 CHAPTER 4 72 Phenotypic evaluation for tolerance to low pH ionp tircal and sub-tropical maize 72 germplasm 4.1 Abstract 72 4.2 Introduction 72 4.3 Materials and methods 74 4.3.1 Description of sites 74 4.3.2 Experimental materials 74 4.3.3 Experimental design 75 4.3.4 Salient management activities 75 4.3.4.1 Fertilizer application 75 vii 4.3.5 Soil characterisation for low pH sites. 75 4.3.5.1 Soil sampling and laboratory analysis 75 4.4 Data analyses 77 4.5 Results 78 4.5.1 Soil analytical results 78 4.5.2 Combined ANOVA for grain yield and agronomtriaci ts at four low pH 79 environments across two seasons 2011/12 and 20 12/13 4.5.3 Estimated contributions to total sum of sqeus aarcross four low pH soil 80 environments for the 2011/12 and 2012/13 seasons 4.5.4 Estimated percent reduction for grain yienld aother salient phenotypic 80 traits at four low pH so iel nvironment versus four optimal environments across 2011/12 and 2012/13 seasons 4.5.5 Genotypic and phenotypic variance compon egnetnse, tic advance and broad8 3 sense heritability estimates across four low pHl esnoviironments combined for 2011/12 and 2012/13 seasons 4.5.6 Mean performance for grain yield and othaeirt st racross four low pH soil 85 environment combined for 2011/12 and 2012/13 sesa son 4.5.7 Pearson’s correlation coefficients betweeanin g yrield and agronomic traits 87 across four low pH soil environments combined f0o1r 12/12 and 2012/13 seasons 4.5.8 Principal component analysis results, eigleunevsa and eigenvectors for the 87 traits across four low pH soil environments combdi nfoer 2011/12 and 2012/13 seasons 4.5.9 Clustering of the maize genotypes evaluatt efodu ar low pH soil 89 environments combined across 2011/12 and 2012/a1s3o nses 4.5.10 Performance of maize genotypes across fpotuimr aol soil environments 91 combined for 2011/12 and 2012/13 seasons 4.5.11 Estimated contributions to total sum of sreqsu a cross four optimal 91 environments combined for 2011/12 and 2012/13 snesa so 4.5.12 Genotypic and phenotypic variance compon, egnetnsetic advance and broad9 4 sense heritability estimates across four optimvailr oenments combined for 2011/12 and 2012/13 seasons 4.5.13 Mean performance for grain yield and othraeirts t across four optimal 95 environments combined for 2011/12 and 2012/13 snesa so 4.5.14 Pearson’s correlation coefficients betwereanin g yield and other agronomic 97 traits across optimal environments combined for1 2/1021 and 2012/13 seasons 4.5.15 Principal component analysis results, eiagleunevs and eigenvectors for the 99 traits across four optimal environments combinerd 2 f0o11/12 and 2012/13 seasons 4.5.16 Clustering of the maize genotypes evaluatt efodu r optimal environments 99 combined for 2011/12 and 2012/13 seasons viii 4.5.17 Combined ANOVA for grain yield and agrono mtriacits for all locations, 101 optimal and low pH for two seasons 2011/12 and 2/1031 2 4.5.18 Estimated contributions to total sum of sreqsu a cross all locations for two 101 seasons 2011/12 and 2012/13 4.5.19 Genotypic and phenotypic variance compon, ebnrotsad sense heritability and1 04 genetic advance estimates across the combinedo ennmveirnts for both 2011/12 and 2012/13 seasons 4.5.20 Mean performance for grain yield and othraeirts t across all environments 105 for 2011/12 and 2012/13 seasons 4.5.21 Pearson’s correlation coefficient betweeanin g yrield and other agronomic 108 traits across optimal and low pH environments conmedb ifor 2011/12 and 2012/13 seasons 4.5.22 Principal component analysis results, eiagleunevs and eigenvectors for the 108 traits across all environments combined for 201 1a/n1d2 2012/13 seasons 4.5.23 Clustering of maize genotypes evaluatedo uart lfow pH and four optimal 109 environments combined for 2011/12 and 2012/13 snesa so 4.6 Discussion 111 4.7 Conclusions and recommendations 115 4.8 References 116 CHAPTER 5 119 Genotype x environment interactions and stabiolitry t rfopical and sub-tropical maize 119 genotypes in Malawi 5.1 Abstract 119 5.2 Introduction 119 5.3 Materials and methods 122 5.4 Data analysis 123 5.4.1 Analysis of variance 123 5.4.2 Stability analysis 123 5.5 Results 123 5.5.1 Analysis of variance for additive main effse mctultiplicative interaction 123 5.5.2 Genotype and GEI scatter biplot and polygioenw vof grain yield across 125 eight environments for 20011/12 and 2012/13 5.5.3 GGE comparison biplot for across optimal laonwd pH sites combined for 129 two seasons 5.5.4 Ranking of genotypes based on both mean ayinedld stability view of the 129 GGE biplot 5.5.5 Cluster analysis of maize genotypes and oenmvirents 129 5.6 Discussion 134 5.7 Conclusions 138 5.8 References 139 CHAPTER 6 143 Evaluation of diallel crosses for combining abi lbiteytween selected tropical and sub-143 tropical maize lines for low pH tolerance ix 6.1 Abstract 143 6.2 Introduction 143 6.3 Materials and methods 146 6.3.1 Experimental materials description 146 6.3.2 Experimental procedures and design 147 6.3.3 Description of sites 147 6.4 Data analysis 148 6.5 Results 148 6.5.1 Performance of diallel crosses 148 6.5.2 Genetic variances, phenotypic variances aenridta hbility estimates for the 149 diallel crosses across optimal and three low pHir oenmv ents in 2011/12 6.5.3 Combining ability and inheritance 154 6.5.3.1 Estimated general combining ability eff efoctrs 12 inbred lines for grain 160 yield and agronomic traits across low pH and oplt iemnavironments in 2011/12 6.5.3.2 Estimated specific combining ability effse fcotr 12 inbred lines for grain 163 yield and agronomic traits across low pH and oplt iemnavironments in 2011/12 6.5.4 Pearson’s correlation coefficients for dila cllreosses between grain yield and1 65 other agronomic traits at optimal environment 6.5.5 Pearson’s correlation coefficients for dila cllreosses between grain yield and1 67 other agronomic traits at low pH environments 6.6 Discussion 167 6.7 Conclusions 172 6.8 References 172 CHAPTER 7 176 General conclusions and recommendations 176 SUMMARY 178 OPSOMMING 180 APPENDICES 182 x LIST OF TABLES Table 1.1 Malawi mean maize hectarage and prodnu ctoiomparisons for 4 2010 versus 2011 and 2011 versus 2012 Table 1.2 Percentage area covered by soil pH v ableuleosw 5.6 in four 6 agricultural development divisions Table 3.1 Description of the tropical and sub-tcroapl imaize genotypes used5 5 in the study Table 3.2 Root length measurements and derived b deafotare and 7 days 62 atter transplanting the glasshouse hydroponic eimxpeenrt Table 3.3 Genetic and phenotypic variances andta hbeilritiy estimates from 64 ANOVA for the measured and derived data Table 3.4 Pearson’s coefficient of correlation amg othne measured and 64 derived data Table 4.1 List of phenotypic and agronomic tranitsd ameasuring procedure 76 Table 4.2 Soil characterization for the low pH s ite 79 Table 4.3 Mean squares for combined ANOVA for g ryaieinld and 81 agronomic traits at four low pH environments ac r2o0s1s1/12 and 2012/13 seasons Table 4.4 Estimated percent contributions to tsoutaml of squares for traits 82 at four low pH environments combined for 2011/12d a2n012/13 seasons Table 4.5 Estimated reduction of grain yield anhde or tsalient traits under 83 low pH versus optimal conditions across the 201 1a/n1d2 2012/13 seasons Table 4.6 Genotypic variances, phenotypic varia nacneds heritability 84 estimates at low pH environments across two sea 2s0o1n1s/12 and 2012/13 Table 4.7 Genotypic coefficient of variation, phteynpoic coefficient of 85 variation and expected genetic advance at low pvHir oennments across two seasons 2011/12 and 2012/13 Table 4.8 Mean performance for grain yield and ro athgeronomic traits 86 across four low pH combined for 2011/12 and 201 2/13 Table 4.9 Pearson’s correlation coefficients foari ng ryield and agronomic 88 traits across four low pH environments for two soenass Table 4.10 Eigenvalues, percentages and cumulpaetirvcee ntages for the 89 measured and derived data across four low pH nsovirl oenments combined for 2011/12 and 2012/13 seasons Table 4.11 Mean squares for combined ANOVA for ng ryaiei ld and 92 agronomic traits at four optimal environments fo0r1 12/12 and 2012/13 seasons Table 4.12 Estimated percent contributions to tsoutaml of squares at four 93 optimal environments across 2011/12 and 2012/1s3o snesa xi Table 4.13 Genotypicσ (2g), phenotypic σ(2p) variances and broad sense 94 (H2b) heritability estimates at four optimal environnmtse across the 2011/12 and 2012/13 seasons Table 4.14 Genotypic coefficient of variation, pohteynpic coefficient of 95 variation and expected genetic advance at optimomalb cined for 2011/12 and 2012/13 seasons Table 4.15 Mean performance for grain yield acrfosusr optimal 96 environments combined for 2011/12 and 2012/13 snesa so Table 4.16 Pearson's correlation coefficients froarin g yield and agronomic 98 traits across all optimal environments for two soenass 2011/12 and 2012/13 Table 4.17 Eigenvalues and eigenvectors for thites taracross four optimal 99 environments combined for 2011/12 and 2012/13 snesa so Table 4.18 Mean squares for combined ANOVA for ng ryaiei ld and 102 agronomic traits for all environments optimal aonwd lpH for two seasons 2011/12 and 2012/13 Table 4.19 Relative percent contribution to toutaml s of squares across two 103 years at eight environments (optimal and low pH) Table 4.20 Genotypic variances, phenotypic varisa nacned heritability 104 estimates across optimal and low pH environmenrt s2 0fo11/12 and 2012/13 seasons Table 4.21 Genotypic coefficient of variation, pohteynpic coefficient of 105 variation and genetic advance across all eightr oenmvients for two years Table 4.22 Mean performance combined across twros yaenad across 106 optimal and low pH environments for 2011/12 and2 2/1031 seasons Table 4.23 Pearson’s correlation coefficients froari ng yield and agronomic 107 traits across optimal and low pH environments f0o1r 12/12 and 2012/13 seasons Table 4.24 Eigenvalues, percentages and cumulpaetirvcee ntages for the 109 measured and derived data across four low pH nsovirl oenments combined for 2011/12 and 2012/13 seasons Table 4.25 Phenotypic and genotypic variances rfaoirn g yield and other 109 traits at four optimal environments Table 5.1 AMMI Analysis of variance for grain yie flodr two years 124 2011/12 and 2012/13 Table 5.2 IPCA1 and IPCA2 scores for the top 20o gtyepnes based on 126 mean grain yield at eight environments for two soenas s Table 5.3 IPCA1 and IPCA2 scores for the eight reonnvmi ents, ranked 126 based on environmental mean for two seasons Table 6.1 Description of 12 maize parental lineesd u isn the diallel crosses 147 and their origin xii Table 6.2 Mean squares for diallel crosses acrpotsims aol and three low pH 149 environments for grain yield and agronomic tranit s2 0i 11/12 Table 6.3 Mean performance of diallel crosses asc oropstimal and low pH 150 environments 2011/12 season Table 6.4 Mean performance of diallel crosses asc trhorsee low pH 151 environments in 2011/12 season Table 6.5 Mean performance of diallel crosses atimt oapl environments in 152 2011/12 Table 6.6 Estimated percent reduction for saliehnetn potypic traits for 153 diallel crosses under low pH versus optimal conodni ti Table 6.7 Genetic variances, phenotypic variannceds h aeritability 153 estimates for the diallel crosses across optimda lt harnee low pH environments in 2011/12 Table 6.8 Combined analysis of variance for GCA SanCdA for diallel 155 crosses for grain yield and other agronomic tracitrso ss optimal and three low pH environments in 2012 Table 6.9 Combined analysis for GCA and SCA forll edli acrosses for 156 grain yield and other agronomic traits across t hlorewe pH environments in 2012 Table 6.10 Mean squares for GCA and SCA effectse ru dnidfferent 157 environments Table 6.11 Relative percent contribution of sums qouf ares for GCA and 158 SCA to total sum of squares across environments Table 6.12 Estimated general combining ability cetfsfe for 12 inbred lines 161 for grain yield and agronomic traits across low apnHd optimal environments in 2011/12 Table 6.13 Pearson’s correlation coefficients u nodpetirmal conditions 16 6 Table 6.14 Pearson’s correlation coefficients aenvde ll of significance under 168 low pH for diallel crosses xiii LIST OF FIGURES Figure 1.1 Map of Malawi depicting eight agricualtul dr evelopment divisions and 3 experimental sites at research stations Figure 1.2 Malawi mean maize hectarage and proodnu cptei r region from 2010 to 5 2012 Figure 3.1 Germination of maize genotypes in nerwinst sp paper and appearance 7 57 days after transplanting Figure 3.2 Partial view of purple colouration anhdo rstened roots observed in 59 susceptible genotypes Figure 3.3 Partial view of new roots emerged fromle rtant genotypes 7 days after 60 transplanting Figure 3.4 Graph of nett seminal root length fonr ogteypes 63 Figure 4.1 Dendrogram based on Euclidean distandc eU aPGMA clustering using 90 morphological data for genotypes at four low pH ireonnvments combined for 2011/12 and 2012/13 seasons Figure 4.2 Dendrogram based on Euclidean distandc eU aPGMA clustering using 100 morphological data for genotypes at four optimavli reonnments combined for 2011/12 and 2012/13 seasons Figure 4.3 Dendrogram based on Euclidean distandc eU aPGMA clustering using 110 morphological data for genotypes at four low pH afonudr optimal environments combined for 2011/12 and 2012/13 snesa so Figure 5.1 AMMI biplot for yield for genotypes anedn vironments across two 127 seasons 2011/12 and 2012/13 Figure 5.2 Genotype and GEI scatter biplot and gpoonly view of grain yield across 128 eight environments for 20011/12 and 2012/13 sea sons Figure 5.3 Genotype and GEI comparison biplot oafi ng ryield across eight 130 environments for 2011/12 and 2012/13 Figure 5.4 Ranking of genotypes based on both myiealdn and stability view of the 131 GGE biplot Figure 5.5 Dendrogram of 45 maize genotypes asa rledv eby UPGMA cluster 132 analysis based on AMMI adjusted mean yields comdb finoer two seasons using Euclidean distance and standard deviatiosnc alsin g method Figure 5.6 Dendrogram of nine environments as rlevde bay UPGMA cluster 133 analysis based on environmental means and Euc liddiestaannce and standard deviation as scaling method Figure 6.1 Dendrogram of 12 maize inbred lines db aosne GCA effects for grain 159 yield across four environments in the 2011/12 sne aso xiv LIST OF APPENDICES Appendix 1 Root length measurements and derivead bdeaftore and 7 days after 182 transplanting a glasshouse hydroponics experiment Appendix 2 Maize genotypes evaluated in the fireialdls t 2011/12 and 2012/13 1 85 Appendix 3 Soil sampling data 186 Appendix 4 Eigenvectors for the measured and dde rdivaeta at low pH 187 environments across two seasons Appendix 5 Eigenvectors from the principal compotn aennalysis for grain yield and 188 agronomic at optimal environments across two sesa son Appendix 6 Soil analytical data interpretation geu id 189 Appendix 7 Mean performance for grain yield acrfosusr optimal environments 191 combined for 2011/12 and 2012/13 seasons Appendix 8 Mean performance for grain yield ande or tahgronomic traits across 193 four low pH environments combined for 2011/12 an0d1 22/13 seasons Appendix 9 Mean performance for grain yield ando angormic traits for low N 195 environment in 2012/13 season Appendix 10 Mean performance combined across twaors y aend across optimal and 197 low pH environment for 2011/12 and 2012/13 seasons Appendix 11 Genotypes used in genotype x environt minetenractions and stability 199 analysis Appendix 12 Estimated specific combining abilityfe ecfts of 12 inbred lines for grain2 00 yield and agronomic traits across low pH and oplt iemnavironments Appendix 13 Mean performance of diallel cross prnoyg aecross optimal and low pH 203 environments in 2011/12 Appendix 14 Mean performances of diallel cross pernoiegs across three low pH 206 environments in 2012 Appendix 15 Mean performance of the diallel crorsosg penies under optimal 209 conditions in 2011/12 Appendix 16 Estimated general combining abilitye ectfsf for 12 inbred lines for 212 grain yield and agronomic traits at low pH enviroennmts in 2011/12 Appendix 17 Estimated specific combining abilityfe ecfts of 12 inbred lines for grain2 13 yield and agronomic traits across low pH environtms e2n011/12 Appendix 18 Estimated general combining abilitye ectfsf for 12 inbred lines for 216 grain yield and agronomic traits for optimal sooiln cditions in 2011/12 Appendix 19 Estimated specific combining abilityfe ecfts of 12 inbred lines for grain2 17 yield and agronomic traits at optimal soil condnitsio 2011/12 xv ABBREVIATIONS AND SYMBOLS AD Days to anthesis ADD Agricultural Development Division AEA Average environmental axis AECa Average environment coordination abscissa AECo Average environment coordination ordination Al Aluminium AMMI Additive main effects and multiplicative inrtaection ANOVA Analysis of variance ASI Anthesis-silking interval ATTC Agricultural Technology Clearing Committee B Boron BKA Baka Research Station BKE Bembeke Research Site BKT Bembeke Turnoff Research Site C Carbon Ca Calcium CIMMYT International Maize and Wheat Improvementn Cteer Cl Chlorine CLA Chitala Research Station cm Centimetre (s) CML CIMMYT maize line CRD Completely Randomised Design Cu Copper CV Coefficient of variation CZE Chitedze Research Station DC Double cross DM Downy mildew DS Days to silking DT Drought tolerant DT2 Distal transition zone ExY Environment by year/season interaction xvi EH Ear height EPP Ears per plot F1 First filial generation FAO Food and Agriculture Organisation of the Udn iNteations FAOSTAT Food and Agriculture Organisation Statis tic Fe Iron FeHEDTA Ferric hydroxethylethylenediaminetriace.t ate FEWSNET Famine early warning system net work FSRL Final seminal root length GxE Genotype by environment interaction GxExY Genotype by environment by year interaction GxY Genotype by year interaction G Genotype GA Genetic advance GCA General combining ability GCV Genotypic coefficient of variation GDP Gross domestic product GEI Genotype by environment interaction GGE Genotype and genotype by environment inteorna cti GLS Gray leaf spot GT Grain texture GWS Genome wide selection GY Grain yield h2b Broad sense heritability H Hydrogen ha Hectare (s) IFPRI International Food Policy Research Insti tute IITA International Institute of Tropical Agriculrtue IPCA Interaction principal component analysis ISRL Initial seminal root length K Potassium KAl (SO4)2 Potassium aluminium sulphate kg ha-1 Kilogram per hectare xvii L Litre LB Leaf blight disease LOX Lipoxygenase LSD Least significant difference LU Lunyangwa Research Station m Metre (s) masl Metre (s) above sea level Max Maximum Mg Magnesium Min Minimum Mn Manganese Mo Molybdenum MOA Ministry of Agriculture MRU Meru Research Station MSE Mean square error MSV Maize streak virus MT Metric ton MVAC Malawi Vulnerability Assessment Committee N Nitrogen NADH Nicotinamideadehyde NBOS National Bureau of Statistics NCSS Number Cruncher Statistical System Ni Nickel NSRL Nett seminal root length NUE Nitrogen use efficiency O2 Oxygen OPV Open-pollinated variety P Phosphorus PAL Phenylalanine ammonia lyase PC Principal component PCA Principal component analysis PCV Phenotypic coefficient of variation PH Plant height xviii Pi Cultivar performance measure POD Peroxidase QPM Quality protein maize QTL Quantitative trait loci r Pearson correlation coefficient R2 Coefficient of determination rcop Cophenetic correlation RDP Rural Development Programme RE Rotten ears RFLP Restriction fragment length polymorphism RL Root lodging ROS Reaction oxygen species Rti Root tolerance index S Sulphur SCA Specific combining ability SE Standard error SH Shelling percentage SL Stem lodging SSA Sub-Saharan Africa SVD Single value decomposition SWT 100 seed weight TSA Tsangano Research Site TSS Total sum of squares UN United Nations UPGMA Unweighted pair-group method with arithmeatvice rages US United States VIG Plant vigour Wi Wricke’s ecovalence WFP World food programme Y Year Zn Zinc Σ Summation σ2e Error variance xix σ2g Genotypic variance σ2i Stability variance σ2o Environmental variance σ2p Phenotypic variance % Percent °C Degrees Celsius xx CHAPTER 1 General introduction 1.1 Origin, importance and production constrainf tmsa oize Maize (Zea maysL .) is an important crop and is favoured as we liln adsispensable food for over one billion people in Sub-Saharan Africa (S SaAn)d Latin America (Guptae t al., 2009). It is a cultivated sub-species of teosian twe,i ld naturally found grass with its centre of origin the Mesoamerican region, now Mexico anedn tCral America (Mangelsdorf, 1974). It was discovered by Columbus’s men in Ciunb a1 492 and later introduced to Europe and Africa by explorers in 1500 as repobrtye dG ibson and Benson (2002). It is a very popular crop but the name “maize” is not Esnhg.l iThe genuZs ea was derived from an old Greek name for a food grass (Mangelsdor7f,4 )1,9 while the sub-speciesm “ays” derived from Spanishm: aíz after Taínom ahiz (Encyclopædia Britannica, 2010). It has a number of uses and in the tropics it is grown fiorer cdt consumption by man and animals as well as various industrial uses (Poweet lal l., 2004). Worldwide, reports indicate that maize is cultivda toen approximately eight million hectares (ha) of low pH soils (Brewbaker, 1985;d Peayn and Gardner, 1992) and yields can be reduced by up to 70% under these conditions c(kWeer let al., 2005). Reports also indicate that on these soils, aluminium (Al) or mganese (Mn) toxicity, calcium (Ca), magnesium (Mg), phosphorus (P) and molybdenum (dMeofi)c iencies are the main causes of yield reduction (Aldriche t al., 1973; Granadoest al., 1993). In Africa, acid soils in the tropical area are estimated to cover 29% of theti nceont (Eswarane t al., 1997). However, von Uexküll and Mutert (1995) reported that low psoHi ls are present all over the world with 41% in America, 26% in Asia, 17% in Africa, %10 in Europe and 6% in Australia and New Zealand. Acidity is a major constraint to ma pizroeduction and other crops on tropical soils. This is because at low pH (pH<5) toxic3 +A iol ns are released into the soil solution, and hinder root growth thus affecting the develonptm oef the entire plant (Kochian, 1995; Kidd and Proctor, 2000). Al toxicity causes shothrti,c k and under developed roots and 1 plants, thus reducing nutrient uptake and incre saussecseptibility to drought (Sasaekti al., 1996). 1.2 Maize production in Malawi It is commonly said that “maize is life” for couinetsr in SSA and this is true for Malawi more than any other country. The National Burea Sut aotfistics (NBOS) of the Government of Malawi data for 2006/07 reported that maize erespernted about 69% of area covered by 16 major crops grown in the country. The FAOSTATr 2fo011 estimated that maize represented about 44% of the total area coveremd obrye than 40 crops in Malawi. Other essential crops include groundnut, tobacco, cas, ssawveaet potato, cotton, rice, soybean, sorghum and millet. Almost 75% of maize in Malaws i ciultivated in pure stands while mixed stands represent about 25%. Cultivation iss tlmy oby resource-challenged smallholder farmers (MOA, 1994). Malawi’s maize requirement is 2.4 million metricn sto (MT) per year and in 2009 the country registered a 1.2 million MT surplus whinle 2i010 the country had a surplus of approximately 800 000 MT this slight reduction oams cpared to the previous year, probably because of drought in some districts in the sount hreegrion (FAOSTAT, 2011). In 2013 the country produced 3.6 MT representing a surplus .o2f M1T (FAOSTAT, 2013). The country saw a record harvest in 2014 of just ov.9e rM 3T (GIEWS, 2015) 1.3 Maize agro-ecology in Malawi Malawi covers an area of 118 000 2k wmhich is relatively small, yet its endowed with diverse agro-ecology areas (Figure 1.1). Aboutm 1i.ll2io n ha are grown to maize which is widely cultivated across the 28 districts which agrreouped into eight Agricultural Development Divisions or ADDs (Karonga, Mzuzu, Knagsu, Salima, Lilongwe, Machinga, Blantyre and Shire Valley) and three ornesg i(northern, central and southern). Approximately 57% of all maize in Malawi is cultitvead in the central region, followed by the southern region (24%) and northern region (1 (9T%ab) le 1.1 and Figure 1.2). Among 2 the ADDs, Karonga, Mzuzu, Kasungu, and Salima conmedb irepresent 80% of all area cultivated to maize in the country (MOA, 1994; MVA, 2C013). Meru Research Site Karonga Baka Research Site Lunyangwa Research Site (Low pH) Kasungu Chitala Research Salima Site Chitedze Research Sites Bembeke Research Machinga Site (Low pH) Tsangano Research Site (Low pH) Bvumbwe Research Site (Low pH) Shire Valley Figure 1.1 Map of Malawi depicting eight agricualtlu drevelopment divisions and experimental sites at research stat ions 3 Table 1.1 Malawi mean maize hectarage and produnc ctioomparisons for 2010 versus 2011 and 2011 v2e0rs1u2s Area (ha) Production (MT) Area (ha) Production (MT) ADD 2009/10 2010/11 % change 2009/10 2010/11 %an cghe 20010/11 2011/12 % change 20010/11 2011/12 ha%n gce Karonga 45855 48960 6.8 116603 137578 18.0 48960 49996 2.1 137578 127381 -7.4 Mzuzu 143569 151262 5.4 307758 357446 16.1 15 1262156046 3.2 357446 344552 -3.6 Kasungu 305909 308921 1.0 752808 804331 6.8 310 892323692 4.8 804331 814454 1.3 Salima 59287 60208 1.6 145859 157168 7.8 60208 6595 6 -5.9 157168 124322 -20.9 Lilongwe 341252 346453 1.5 714180 784013 9.8 35436 4 347140 0.2 784013 739271 -5.7 Machinga 287509 287976 0.2 389779 382004 -2.0 97268 7 276207 -4.1 382004 296374 -22.4 Blantyre 250026 254663 1.9 332359 542210 63.1 66235 4 256199 0.6 542210 438895 -19.1 Shire Valley 42211 42106 -0.2 42211 28594 -32.3 2 1046 31890 -24.3 28594 20743 -27.5 Total 1475618 1500549 18.0 2801557 3193344 87.4 15 00549 1497829 -23.4 3193344 2905992 -105.3 Source: Ministry of Agriculture Crop Estimates2 f0o1r 0, 2011, 2012 4 2500000 2000000 1500000 1000000 Ha 500000 Production 0 (MT) Figure 1.2 Malawi mean maize hectarage and prodounc ptier region from 2010 to 2012 1.4 Abiotic constraints to maize production in wMia la Soil acidity is prevalent in most parts of Malawnid a a limiting factor in crop production. The increasing population is creating pressurea onnd land continuous mono-cropping and slashing and burning of crop residues during lalneda rcing have exacerbated the problem. More acid soils are found in the high rainfall asr e(>a1000 mm per year) where there is moderate to high leaching, while the alkaline s aoriles found in low rainfall areas (< 500 mm per year). Regions with soil pH less than 5.v5e h baeen identified in the country and according to the soils database prepared by thles CSoimmodity Team, over 40% of the country has such soil. The largest hectarage oyf avceird soils are found in the following ADDs: Lilongwe, Mzuzu and Blantyre. Chilimba (199 r4e)ported higher Al saturation percentages in some areas of Bembeke, Lunyangwhaa,t aNbkay and Mulanje. The soil pH in the ADDs in the country is outlined in Table .1 .2 5 Table 1.2 Percentage area covered by soil pH v ableuleosw 5.6 in four agricultural development divisions ADD Soil pH Area % coverage Blantyre 4.2 – 5.5 36 Kasungu 4.2 – 5.5 10 Lilongwe 4.7 – 5.5 65 Mzuzu 4.4 – 5.5 33 Source: Chilimba and Saka 1998 The well-known low pH soils are found in most p aorft sBembeke, Kanyama and Mayani in Dedza; Namwera rural development programme (R iDnP M) angochi; Tsangano in Ntcheu; Mulanje RDP in Mulanje; Thyolo RDP in Thoy;o lNkhata Bay RDP in Mzuzu ADD (Lunyangwa, Ntchenechena, Mphompha, Uzumalau, zMuz city, Mzimba central and South Mzimba) and Misuku Hills in Chitipa. High psHo ils or alkaline/sodic soils are located in Shire Valley, along Lake Chilwa and L aMkealawi (Chilimba and Komwa, 2003). In such low pH soils, crop yields are limdi taend sometimes total crop losses occur. For instance, Munthali and Chilimba (2004) repo rate ydield reduction of more than 85% in low pH soils in Lunyangwa as compared to thee nptoiat l yield of 8.5-10 ton h-1a for maize hybrids under normal fertility conditions. The problem of low-soil pH can be solved by usinogil samendments such as liming, although most farmers in developing countries cat nanfford such amendments (Pandeet y al., 1994). A more sustainable solution would be toe cste Al l tolerant maize genotypes for use in acid soils which, in the long run, is lesxsp eensive, sustainable and more environmentally friendly. Other abiotic stresses are droughts and floods coonm imn low-land areas of the coun.t ry Mazunda and Droppelmann (2012) reported that ionu an tcry of which its economic base is heavily dependent on agriculture, not only ahree rtural livelihoods affected due to the negative impacts on the agricultural sector, bunt- fnaorm and urban households are not spared either, given the strong relationship ofd upcrotion and prices between agriculture and the rest of the economy. According to the Mai laVwulnerability Assessment Committee (MVAC, 2010), 718 000 people were decdl afroeod insecure between March and June in eight districts in southern Malawi dtou ep oor harvests as a result of prolonged 6 dry spells in the 2009/2010 season. The numbefrf eocf taed people is expected to increase to 1.1 million by October 2010 (FEWSNET, 2010). FSENWET (2012) estimated that above one and a half million people would be ind n oefe food relief between October 2012 and March 2013. Flooding affected the country in early 2013 in s uac whay that the United Nations (UN) World Food Programme (WFP) in conjunction with tGheo vernment of Malawi were providing food relief to about 6 700 households cwhh wi ere flood victims (FEWSNET, 2013). Incidences of food shortages worsen andp sphraicre increases occur which reduce households’ disposable incomes. It is mostly smscaalll-e farmers and those residing in the flood-prone southern regions of the country thaty svtulnerable (Selka, 2012). The economic losses as a result of climate relateds tdeirssa are evident: Malawi loses 1.7% of its gross domestic product (GDP) on average evearyr ydue to the combined effects of droughts and floods. This is equivalent to approaxteimly US$22 million in 2005 prices (Mazunda and Droppelmann, 2012). 1.5 Biotic constraints to maize production in Mia law Economic importance maize diseases in Malawi inec lvuidral disease such as maize streak virus (MSV), fungal diseases such as leaf blighBt )( Lcaused byE xserohilum turcicum (Leonard and Snugs) and gray leaf spot (GLS) ca buys ethde pathoge nCercospora zeae- maydis (Tehon and Daniels) and downy mildew (DM) anotfhuenrg al disease caused by the genusP eronosclerospor.a GLS can cause yield losses of up to 60% (Ringnedr a Grybauskas, 1995). The most destructive diseaseld -woidre is DM (Frederiksen and Renfro, 1977) and in Malawi two species are knoow nc atuse this disease, these Pa.r e philippinensis andP . sorgh.i Two pathotypes oPf . sorgh ihave been reported, one capable of infecting both maize and sorghum and the othneferc iting only maize (Anasoet al., 1987). The diseas weas first identified in maize in Mozambique (PlumDbh-indsa and Mondjane, 1984). In Malawi its occurrence on sormgh wuas reported by Beck (1980) and its observation on maize was in the 2004/05 seains othne Blantyre ADD where over 40 000 farm families were left food insecure espeyc iainll the Mulanje and Thyolo districts. Adenle and Cardwell (2000) reported that the ta bsrsaeclts may proliferate, resulting in a very bushy appearance, causes distortion and/notri nsgtu of the maize plant. It frequently 7 occurs in areas of fields subject to flooding wh tehree zoospores infect the growing point of the young maize plants. Another biotic stress in maize production in Mal aisw wi itch weedS triga spp .Its origin is not very clear (Holme t al., 1977) and it is believed to be indigenous top itcraol and sub- tropical Africa and Asia. In Malawi the most impaonrt genera for cereals Sis. asiatica locally known ask aufiti and is the most widely spread in the country apso osepd to other witch weeds likeS . hemonthic aand Alectra vogell ifor legumes. Kabambet al. (2008) reported that yield losses depend on level of tinafteiosn, susceptibility of the maize genotype, soil fertility and crop management prcaecst.i Striga seeds are shed in large numbers (over 50 000 per plant) and remain viaobrl elo fng time (up to 20 years) (Ramaiah et al., 1983). 1.6 Malawi National Maize Breeding Programme The Malawian National Maize Breeding Programme,h witist main office at Chitedze Agricultural Research Station, was established wthiteh aim of variet ydevelopment and breeder seed production as well as seed distrinb utoti ogrowers. Major achievements have been reached in the development of new maize vieasr ieatnd identification of improved varieties for tolerance to stresses obtained frothme ro breeding institutions. To this effect nine maize hybrids were released in 2013, threweh oicfh are both drought and low nitrogen (N) tolerant (CIMMYT, 2013). For drought alone, tphreogramme has released a total of five cultivars since 2009. These are MusungabaZnMja 3(09), Mwayi open-pollinated variety (OPV) ZM523, MH30, MH31 and MH32. In termofs dissemination, four newly released hybrids of 2013 were already selectepdr foodru ction by different seed companies. In terms of nutrition, two quality protein maize P(QM) varieties were released in 2008 and 2009, an OPV, Chitedze2QPM and a hybrid (MH29)p, erecstively (Kaonga, 2009; Mviha et al., 2011). There are a good number of released dhsy bfrroi m the programme which are tolerant to GLS, MSV as well as LB and are disttreibdu by seed companies. 8 Despite all these achievements varieties for low toplHerance are yet to be developed. Hence the objectives of this study we re: 1. To evaluate maize genotypes of diverse genaertica bvility for tolerance to Al as a proxy for low pH tolerance. 2. To study maize genotypes of diverse genetica vbailirtiy for tolerance to low pH soils by use of phenotypic and morphological traits. 3. To study the genotype by environment interac (tGioxnE) and stability of the tropical and sub-tropical maize genotypes. 4. To estimate combining ability among well adap intebdred lines and low pH lines from CIMMYT-Colombia. 1.7 References Adenle, V.O. and K.F. Cardwell. 2000. Seed transimonis of Peronosclerospora sorg,h i causal agent of maize downy mildew in Nigeria. Pt Plaanthology 49: 628-635. Aldrich, S.R. 1973. Plant analysis: Problems anpdo ortpunities. pp. 213-222. I nL:.M. Walsh and J.D. Beaton (eds.) Soil testing and palannatlysis. SSSA, Madison, WI. Aldrich, S.R., W.O. Scott and E.R. Leng. 1975. Mrond ceorn production. A & L Publication Champaign, IL. Anaso, A.B., P.D. Tyagi, A.M. Emechebe and S.K. Mzoa.n 1987. Identity of a downy mildew in Nigeria Guinea savannah. Samaru Jourfn Aagl oricultural Research 5: 13-22. Beck, B.D.A. 1980. Sorghum diseases in Malawi.R I.nJ:. Williams, R.A. Frederiksen, L.K. Mughogho (Eds.). Proceedings of the Internnaatilo Workshop on Sorghum Diseases, 11–5 December 1978, Hyderabad, Indiar.a A Pnrdadesh, India: International Crops Research Institute for the S-Aermidi Tropics. Brewbaker, J.L. 1985. The tropical environment mfoari ze cultivation. In: A. Brandolini, F. Salamini (Eds.). Breeding strategies for mairzoed upction improvement in the tropics. Firenze: FAO; Instituto Agronomico per lIt’rOemare. Chilimba, A.D.C. 1994. Annual Report: Soil Fertyil itand Plant Nutrition Commodity Team, Department of Agricultural Research Serv icMesin,istry of Agriculture, Lilongwe, Malawi. 9 Chilimba, A.D.C., S. Chigwenembe, B.W. Lungu anAd .P S.onjera. 2004. The effects of different organic fertilizers and their interactsio wn ith inorganic fertilizer on maize yield. In: Annual report of the Soils Fertility a nPdlant Nutrition Commodity Team, Ministry of Agriculture, Lilongwe, Malawi. Chilimba, A.D.C. and M.M. Komwa. 2003. Soil fertyil istatus in Lilongwe Agricultural Development Division. A Final Report. Chitedze Acgurlitural Research Station, Lilongwe, Malawi. CIMMYT. 2013. Drought Tolerant Maize in Malawi: Ar iBght Spot for DTMA. DT Maize 2: 1-4. Encyclopædia Britannica, 2010. The Teacher – Flryie gnudide to the evolution of maize. Encyclohttp://maize.teacherfriendlyguide.org/indpehxp./what-is-maize. Eswaran, H., Reich, P., and Beigroth, F. 1997. Gall odbistribution of soils with acidity. In A. C. Moniz et al. (Eds.)S, oil Plant Interactions at Low pH(p p. 159-164). Brazilian Soil Science Society Vicosa. FAO. 2001. Food balance sheets: A handb. oRookme. ftp://ftp.fao.org/ docrep/ fao /011 /x9892e/x9892e00.pdf FAOSTAT. 2011. Food and Agriculture Organizationa tSisttical Database: http// faostat. fao.org. FAOSTAT. 2013. Statistical Database of the Food Aagnrdiculture of the United Nations. http://www.fao.org FEWSNET. 2007. Farming and Early Warning System wNoerkt. Summary of crop production estimates, Lilongwe, Malawi. FEWSNET. 2010. Malawi Food Security Outlook Upd aFtEeW. SNET Malawi, Lilongwe. FEWSNET. 2012. Malawi Food Security Outlook Upd aFtEeW. SNET Malawi, Lilongwe. FEWSNET. 2013. Malawi Food Security Outlook Upd aFtEeW. SNET Malawi, Lilongwe Frederiksen, R.A. and B.L. Renfro. 1977. Globatlu sst aof maize downy mildew. Annual Review of Phytopathology 15: 249-275. Gibson, L. and G. Benson. 2002. Origin, history aunseds of cornZ (ea mays.) Iowa State University. Department of Agronom. y Giews (Global Information and early warning systeomn food and agriculture). 2015. Country Briefs Malawi. http://www.fao.org/giews/cnoturybrief/country.jsp Granados, G., S. Pandey and H. Ceballos. 1993.o Rnsees pto selection for tolerance to acid soils in a tropical maize population. Crop Scie3n3ce: 936–940 . 10 Gupta, H.S., P.K. Aggarwal, V. Mahajan, G.S. Bi sAh.t , Kumar, P. Verma and A. Srivastava. 2009. Quality protein maize for nuotrnitai l security: Rapid development of short duration hybrids through molecular markaesrs isted breeding. Current Science 96: 230–237. Holm, L.G., D.L. Plucknett, J.V. Pancho and J.P.r bHeerger. 1977. The world’s worst weeds: Distribution and Biology. The University Press oaf wHaii, Honolulu. pp 456– 464 Kabambe, V.H., P. Ngwira and R.P. Ganunga. 200t8e.g rInated Management of Witch weed (Striga asiatica) in Malawi. Lilongwe, Malawi. Kaonga, K.K.E. 2009. Release dossier for Chited QzeP 2M and MH29 QPM presented to the Agricultural Technology Clearing Committee (ACT)C. 17th February 2009. Lilongwe. Malawi. Kochian, L.V. 1995. Cellular mechanisms of alummin iutoxicity and resistance in plan ts. Annual Review of Plant Physiology and Plant Moleacr uBl iology 46: 237-260. Mangelsdorf, P.C. 1974. Corn: its origin, evolut aionnd improvement. Harvard Univ. Press, Cambridge. pp. 3-10. Mazunda, J and K. Droppelmann. 2012. Maize Consiuomn pEt stimation and Dietary Diversity Assessment Methods in Malawi. Lilongwe,a lMawi. International Food Policy Research Institute (IFPRI) Series numbe.r : 11 MOA (Malawi Government Ministry of Agriculture) 1949. Guide to Agricultural production in Malawi 1994. Extension Aids Branch.i nMistry of Agriculture. Lilongwe. Munthali, M.W. and A.D.C. Chilimba. 2004. Effectfs c oomposting materials and methods of composting on quality of compost and maize ysi.e lIdn: 2004 Annual Report of Soils and Agricultural Engineering Commodity Gro uMpi,nistry of Agriculture, Chitedze Research Station, Lilongwe, Malawi. pp. . 52 MVAC (Malawi Vulnerable Assessment Committee). 2.0 1F0ood Security Monitoring report. Lilongwe, Malawi. MVAC (Malawi Vulnerability Assessment Committee)0.1 23 National Food Security Forecast, April 2013 to March 2014. Bulletin No1. 39 /Volume 1. 11 Mviha, P.J.Z., A.P. Mtukuso, M.H.P. Banda and BC.Fh.is ama. 2011. A Catalogue of Agricultural Technologies released by the Minisotrf yA griculture and Food Security 2004-2011. Department of Agricultural Raersceh Services. Lilongwe, Malawi. pp.4-5. Nordin, S. 2005. Low Input Food and Nutrition Seitcyu: rgrowing and eating more using less. Malawi: World Food Programme. Pandey, S., H. Ceballos, R. Magnavacca, A.F.C. aB-Fahilhi o and J. Duque-Vargas. 1994. Genetics of Tolerance to Soil Acidity in Tropical aMize. Crop Science 34: 1511–1514. Pandey, S. and C.O. Gardener. 1992. Recurrent tisoenl efcor population, variety, and hybrid improvement in maize. Advances in Agronom8y: 14-87. Plumb-Dhindsa, P a.nd A.M. Mondjane. 198. 4Index of plant diseases and associated organisms of Mozambiqu. eTropical Pest Management 30: 407-4 29. Powell, J.M.R., A. Pearson and P.H. Hiernaux. 2 0R0e4v. iew and Interpretation. Crop livestock interactions in the West Africa Dry lan. dAsgronomy Journal. 96: 469-483. Kidd, P.S. and J. Proctor. 2000. Effect of Alummini uon the growth and mineral composition ofB etula pendul aRoth. Journal of Experimental Botany 51: 1057-1. 066 Ramaiah, K.V.C., C. Parker, M.J.V. Rao and L.J. sMeulms an. 1983. Striga identification and control handbook. Information Bulletin No. 1In5t,e rnational Crops Institute for the Semi-Arid Tropics, Patancheru, A.P., India. Ringer, C.E. and A.P. Grybauskas. 1995. Infectiyocnle c components and disease progress of grey leaf spot on field cover. Planste Daise 79:24-2. 8 Sasaki, M., Y. Yamamoto and H. Matsumoto. 1996. niLnig deposition induced by aluminium in wheat T(riticum aestivum) roots. Physiologia Plantarum 96: 193–198. Selka, P.C. 2012. Malawi: Resilience in the Fac eP eorfsistent drought. US Agency for international Development. http://www.usaid.gov/wt-hwae-do/working-crises-and- conflict/building-resilience/malawi-2012 Von Uexküll, H.R. and E. Mutert. 1995. Global extt,e dnevelopment and economic impact of acid soils. In: R.A. Date, N.J. Grundon, G.E.y mRaet and M.E. Probert (Eds.). Plant-Soil Interactions at Low pH: Principles anda nMagement, Dordrecht, The Netherlands, Kluwer Academic, pp. 5-19. 12 Welcker, C., C.B. Andreau, C. De Leon, S.N. Parnein, tJo. Bernal, J. Felicite, C. Zonkeng, F. Salazar, L. Narro, A. Charcosset and W.J. H o2r0s0t.5. Heterosis and Combining Ability for Maize Adaptation to Tropical Acid Soi:l sImplications for Future Breeding Strategies. Crop Science 45: 2405-2413. 13 CHAPTER 2 Literature review 2.1 Importance of maize and consumption levels Maize is the most significant cereal crop in thea mGirneae family in eastern and southern Africa, representing over 29% of the total harvde satreea of annual food crops and 25% of total caloric intake and income (FAOSTAT, 2010 )i.s I tone of the most important food staples in SSA, providing nourishment to over 30il0li omn resource-poor smallholders. Its cultivation spans the entire continent and it ies dthominant cereal food crop in many countries accounting for 56% of the total harve satreeda of food crops and 30-70% of the total caloric consumption (FAOSTAT, 2007). Consumption is high in Southern Africa; the per itcaa paverage is about 195 kg in South Africa, 181 kg in Malawi, 168 kg in Zambia and 1k5g3 in Zimbabwe (Hassan, 1998). According to Calbae t al. (2001) it was estimated that for SSA to be foodu rse cby 2050, food production should be multiplied by seven ams pcaored to the 1995 level. This requires proper planning for increased agricultural prodvuitcyt iwhich is sustainable without or with minimal environmental degradation. 2.1.1 Important abiotic factors affecting maized purcotion The major abiotic constraints to maize productinocnl uides drought, low N soils and low pH soils. With respect to low pH, maize is planotend a pproximately 8 million ha of acid soils all over the world (Brewbaker, 1985; Pandnedy Gaardner, 1992). Soil acidity is found to be a major yield-limiting factor for many croapns d covers extensive areas of land in tropical, sub-tropical and temperate zones; witwh ploH occupying approximately 3.95 billion ha, about 30% of the ice free land of thoer lwd (von Uexküll and Mutert, 1995). The lower yield of crops grown in acid soil is basicya dllue to combinations of low pH, toxicity of iron (Fe), Al and Mn as well as deficienciesN o,f P, Mg and Ca. However, Al toxicity was found to be the main problem in maize produnc btieocause of root growth inhibition, consequently reducing the water and nutrient up tankde its interference in different 14 physiological processes of crop development (Reto ayl ., 1988). The key effect of low pH soil on the plant is a slow growing root systemc, oamcpanied by the establishment of surface roots. This negatively influences the ufs seo oil nutrients and induces plants to be more susceptible to drought (Piñereot sa l., 2005; Hartwige t al., 2007). Soil amelioration can be implemented by correcting the low pH soailt usst. However, the use of soil amendments such as liming, which is well known ntcor eiase soil pH, may have some adverse effects to the environment and have a treamryp oeffect and are too expensive for resource challenged farmers in developing coun. tTrihees low pH change has been reported to occur only in a restricted top soil layer upoimni nl g, while the sub-soil surface layers of the soil profile with toxic Al remain acidic (Cusdtióo et al., 2002). 2.2 Concept of low pH, definition and origin The concept of low pH first came about by a Dancihsehm ist, Soren Peder Lauritz Sorensen in 1909. Soil pH is a measure of the acidity ori cbitays in soils and pH is defined as the negative logarithm of hydrogen ions +( Hor, more precisely, 3HO+ aq) in a solution. According to Brady (1990) the pH scale ranges fr0o mto 14, with 7 being neutral. According to this notion, a pH below 7 denotes iatyc iadnd above 7 denotes alkalinity. Soil pH is considered a significant variable in soil sit adsictates many chemical processes that take place. It significantly affects plant nutri eanvtailability by determining the chemical forms of the nutrient. The optimum pH range for mt polasnt species is between 5.5 and 7.0 however, some plants have adapted to thrive ata pluHe vs beyond this range. Acid soils have a low pH because of the parent rmiaalste from which they derived or originated from through weathering and have lowi cb caastion (Ca, Mg, K and Na) content because these elements have been reduced fromoi lt hbey lseaching or via harvested crops (Granadose t al., 1993). Generally acid soils have low pH and conn toaixic levels of Al and Mn also are deficient in Ca, Mg, P, K, and MDou q(ue-Vargase t al., 1994) and occurring mainly in the form of stable Al silicacteo mplexes, which is non-toxic to plants (Ma and Ryan, 2010). When Al solubilises and foormctsa hedral hexahydrate [Al( 3+2HO)6] also known as A3l+, it becomes toxic to plants even at micro-molar ceonntrcation (Kochian et al., 2005). 15 Globally, 30% of all land area is reported to bem cporised of low pH soils and 50% of the world’s cultivated lands are potentially acidicu; sth Al toxicity is considered as one of the most significant limitations to crop production ñ(Periose t al., 2005). In Brazil, more than 500 million ha are reported to have acid soils,e ecisaplly those covered by Savannah (Cerrado biome) vegetation (Vitorelleot al., 2005). The soils of these areas have high acidity (average pH 4.6), a high concentration lo af nAd Mn, and deficiencies of C2+a, Mg2+ as well as P. These limitations, if not correctceadn, lead to remarkable reduction in crop productivity. Development of genotypes toleranlto two -soil pH has gained importance in recent years. There is great variability in lowl- spoHi tolerance between species and even between genotypes within species (Hueatn agl ., 2009). The mechanisms of tolerance to Al can be summarised into two classes: (i) thosee tlhimatin ate absorbed Al or prevent/reduce its uptake by the roots (Al exclusion) and (ii) odxeiftication mechanisms, which usually act by Al complexation, followed by the transfer ando rsatge of these complexes in vacuoles (internal tolerance) (Hartwiegt al., 2007). According to Kochiane t al. (2005), the main site of Al accumulation and ctoitxyi is the root meristem, primarily the distal part of the ntsraition zone. The rapid root growth inhibition after exposure signifies that the Al tiannstly terminates cell enlargement and elongation before interfering with cell division o(Kchian et al., 2005). After an adequate exposure of the root system to Al, its toxicitym isa nifested through a set of symptoms expressed in its continuous and increasing effne ctht eo morphology and physiology of the roots, which involves decrease in the followingo: mbai ss; the number and length of the roots, often coupled with an increase in the meaadniu rs and root volume; and the uptake of water and mineral nutrients, resulting in se vleorseses of root elongation and the subsequent productivity. 2.2.1 Research findings on aluminium toxicity tesf fec Studies showed that the binding of Al to cell wcaolml ponents changes the cation exchange capacity (Pandae t al., 2009). Ma et al. (2004) reported that visco-elasticity and other properties of the cell wall are affected, resul tiin galterations that interfere with growth. Al can cause decline in the elasticity of the cwealll l and stimulate the synthesis and accumulation of lignin (Peixoteot al., 2007) through the activation of a peroxidase (P OD) 16 linked to the cell wall, which is involved in them pi rovement of hydroxyproline-rich glycoprotein binding to phenolic acids. The enzy maectsivated by Al are comprised of nicotinamideadehyde (NADH) oxidase, phenylalaninme maonia lyase (PAL), and lipoxygenase (LOX). NADH oxidases are responsiboler tfhe synthesis of hydrogen peroxide, which is significa nfot r rapid polymer binding catalysed by the cell lw PaOl D. PAL is the key enzyme in the biosynthesis of phoepnryolpanoids and LOX is responsible for the peroxidation of membrane polyunsaturatettdy facids resulting in the formation of hydroperoxides. These compounds are reported htoig bhely reactive and quickly degraded into compounds that, by the octadecanoic pathwsauyl trse in the production of jasmonic acid, which functions in the lignin synthesis silglinag pathway (Xuee t al., 2008). Kochian et al. (2004a; 2004b) indicated that Al can disrupt cthyeto skeletal dynamics, interacting with microtubules and actin filamenAtsl. can also interfere with signal transduction, particularly in the C2+a signalling pathway (Rengel and Zhang, 2003). According to Sivagurue t al. (2000) and Joneest al. (2006) Al can increase callus synthesis, blocking the plasmodesmata and preventing cell wloaollsening, thus limiting the expansion of cells. The plasma membrane has a ivnelgya ctharged surface, making it a sensitive target for Al toxicity. Al strongly bin dtso phospholipids, which leads to alterations of the lipid composition (Peixoetto a l., 2001), decreases membrane fluidity and increases the folding of density of lipids (Cheetn a l., 1991a; 1991b). Al can also inhibit the H+- ATPase in the plasma membrane, which deters tehvee lodpment of and maintenance of the +H gradient (Ahn et al., 2001). Therefore Al interferes with transportation of secondary ions, indirectly cagu scinhanges of ion homeostasis in root cells. Al also rapidly and effectively inhibits t hineflux of Ca2+ into cells by modulating the activity of transporters which causes alterationn tsh ei membrane potential (Kochieatn a l., 2005). It has strong affinity for phosphate grouthpast makes the A3+l bind to DNA, negatively affecting its template activity and cmhraotin structures (Silvae t al., 2000) and this alters the cell division process (Barceló Panodsc henrieder, 2002; Kochieatn a l., 2005). 2.3 Mechanisms for low pH tolerance It is important to note that plants have developvaerdio us mechanisms to overcome the effects of toxic Al in the soil. These mechanismasn cbe divided into two groups (i) 17 symplastic mechanisms comprising of immobilisatoiorn n eutralisation of Al within the cell and (ii) exclusion or apoplastic mechanismast tdheter the Al from penetrating into the cell, by its immobilisation or neutralisation ine t hrhizosphere (Kochian, 1995; Samac and Tesfaye, 2003). In the symplastic mechanism, Ahl iwn itthe cell is reported to react with several entities such that it can form complexetsh woriganic acids (Foy, 1988; Taylor, 1988), with proteins or other compounds (Suhaydda Haanug, 1995). Internal Al is kept inactive in the cytoplasm or in the vacuoles; tihs isa n advantage because it prevents its negative effects in many cellular processes. Howr, etvhe intracellular mechanisms of tolerance are not well understood, since both atonlte rand sensitive plants have an accumulation of Al when grown in soil conditions hoifgh availability of this element. Different forms of Al can be transported into valceuso, where it is stored without causing further damage to the cell. The exclusion mechasn iosfm Al are well studied (Samac and Tesfaye, 2003; Kochiane t al., 2004a; 2004b) and validated on the basis of tgice, ne physiological and molecular evidence. In these maneicshms, chelating compounds are reported to be released by the roots forming noxnic- tcoompounds with Al, avoiding the entry of this element into cells. In a number of crop species, the exudation of oicrg acnids by root apices is a major means of Al tolerance as reported in maize (Piñeerot sa l., 2002), wheat (Sasaekti al., 2004), and sorghum (Magalhaeest al., 2007). On the other hand, organic acids, eslplye ciiatrate and malate, are reported to form stable complexes twhieth A l3+ in the rhizosphere, reducing the toxic effects in the root system (Kocheiatn a l., 2004a; 2004b). 2.3.1 Genes and inheritance for tolerance to aliuumi ntoxicity Genes play a significant role in Al tolerance. Tfihrset gene identified for Al tolerance isolated in plants was thAeL MTI gene in wheat which is a malate transporter whsic h i activated by Al (Sasakei t al., 2004). GenesS bMATE (Magalhaese t al., 2007) and HvMATE (Furukawae t al., 2007) were isolated from sorghum and barleye rcetsivpely and function as a citrate transporters, also induce dA lb. yAbout two years later, advances in research led to the identification of homologounse gse of theA LMT and MATE multigene families which were isolated from several othern pt lsapecies. In addition, a transcription factor of the zinc finger type calleSdT OP,I is related to Al tolerance in Arabidopsis, which 18 functions in regulating the expression AotfM ATE and AtALMT genes (Liue t al., 2009). Recently,N ramp aluminium transporter 1 (Nrat,1 e)xpressed in the plasma membrane and in the tonoplast, was identified to be associateithd Awl tolerance in rice (Xiae t al., 2010) and this suggested a possibility of involvementh w thite flux of Al and its mobilisation to the cell vacuole. It is important to note that the genetic contro l Aol ftolerance in crops varies from an inheritance controlled by one or two genes, as rovbesde in wheat, to a quantitative inheritance, where genes with smaller effects sa cmt oadifiers, such as in maize (Cançado et al., 2005; Ferreirae t al., 2006;). In wheat, tolerance to Al appears to bnet rcoolled by one or two major genes, with the main gene locaotne dc hromosome 4D (Aniol and Gustafson, 1984; Lagoest al., 1991). Delhaize t al. (1993) associated the Alt1 locus with a large proportion of the variability in toleranacme ong wheat cultivars. Subsequently, the ALMT1 gene which encodes a malate transporter activbayt eAdl, was cloned by Sasaekti al. (2004) and would be the gene underlying the Altc1u slo. Minella and Sorrells (1992) reported that simplhee irnitance of Al tolerance was observed in barley and identified a genAe lp( ) which had a major effect in Al tolerance (Mine allnad Sorrells, 1992; 1997). They concluded that the avtiaorni s in Al tolerance among barley cultivars were controlled by different alleles ahtis t locus; however, other genes with smaller effects may have an influence on this . trTaihte Alp gene was mapped to chromosome 4H (Tanegt al., 2000). 2.3.2 Genetic variability in various crops for ainluiumm tolerance Different crop species exhibit different behavio uinrs soils with high Al saturation (Parentonie t al., 2001). Variations exist within crop species atrnibde s. For instance rye is considered to be the most Al tolerant species oef Tthriticale tribe (Miftahudin and Gustafson, 2002) and genes with larger effects lo tonl eArance were identified to be located on chromosomes 6RSA l(t), 3RL (Alt2) and 4RL A( lt3). Parentonie t al. (2001) reported some species considered to be meexltyr tolerant to Al: some of the tropical forage grass species (Ga, mSibganal, Jaragua and Capitata grass), 19 cassav aand cowpea. Crops such as rice, coffee, potatob,e ru, bpalm oil, rye and oat are considered to be highly tolerant to Al. However,e sewt potato, maize, cabbage, wheat, millet, pea, eggplant, soybean, elephant graslse, yb, aornions, beet, pumpkin, sorghum and shrubs like leucaena present low to medium toleer atnoc Al. Carrots, spinach, celery, cotton, common bean and alfalfa are among the essp ewchi ich are extremely sensitive to Al. A large degree of interspecific variability Aofl tolerance was reported in several crop species (Parentoneit al., 2001; Samac and Tesfaye, 2003). Also Caneiat toa l. (2007) indicated a wide morphological variability for Aoll et rance in a group of 13 sorghum lines, ranging from highly sensitive (20% relative rooot wgrth) to highly tolerant (>100% relative root growth), when measured in nutrient solutiont acoining 60 µM of Al activity. Al tolerance in maize is reported to be of compilnehxe ritance, since progenies derived from crosses between tolerant and sensitive lihnoews scontinuous frequency distributions under Al stress (Magnavaceat al., 1987; Sawasaki and Furlani, 1987). Magnaveatc al . (1987) reported a predominance of additive effeinc tsh e genetic variation linked with Al tolerance in maize. However, Mooent al. (1997) identified a gene with partial dominance (ALM1) responsible for tolerance to Al toxicity in thsisp ecies with the favourable allele identified in a line derived from a somaclonal vaanrti of the cateto race (Cat-100-6). Subsequently, Siboevt al. (1999) mapped two quantitative trait loci (QTLcsa),ll ed ALM1 and ALM2 on chromosome 6 and 10, respectively, hw haicre involved in the genetic control of Al tolerance in maize. Ninamango-Cárdse enta al. (2003) also mapped five Al tolerance QTLs on maize chromosomes 2, 6 and 8la, ienxinpg about 60% of the phenotypic variation of the trait. Of the two QTLs (ALM1 andL AM2), only the one on chromosome 6 was consistent between the two genetic mappiundgi esst in maize, being equivalent to ALM1. Recently, Marone t al. (2009) characterised a member of the MATE famnil y i maize, Zm MATE1, co-localised with the major Al etoral nce QTL in the same region as chromosome 6. According to the author this candei dgaetne encodes a protein located in the plasma membrane that activates the citratea sre lien the root apex. Al tolerance in rice appears to be quantitativenlhye irited (Khatiwadae t al., 1996; Wue t al., 1997). This observation was confirmed by mappsitnugd ies in crosses between different crop species of the genus (Wetu a l., 2000; Nguyene t al., 2002). The evaluation of diallel crosses in soybean revealed that additive efferect sp raedominant in Al tolerance (Spehar, 20 1995; Spehar and Galwey, 1996). Bianchi-Heat lal l. (2000) reported more than five QTLs with minor effects, and concluded that the contorof l Al tolerance in this species is quantitative. The diallel crossing in alfalfa, ao pc rhighly Al sensitive, also concluded that non-additive effects were more important than tdhdei taive effects in the control of Al tolerance (Campbeellt al., 1994). 2.4 Types of mechanisms for aluminium tolerance Plants have developed different means to overcoml set rAess either by precluding 3A+ l from entering the root (extrusion mechanisms) o br ebiyng able to deactivate or neutralise toxic Al3+ absorbed by the root system which is a true tnocleer amechanisms. So far the only well documented mechanism of Al resistancteh eis e xclusion of Al from the root tip based on exudation of organic acids, which cheAlalt3e+ creating stable, non-toxic complexes (Kochiane t al., 2004a; 2004b). The root apex was identified as the main site o-ifn Adluced root growth inhibition (Bennet and Breen, 1991; Ryaent al., 1993). The most frequently measured effect o tfo Axilcity is the inhibition of root growth, but it is importantot bear in mind that a number of physiological and biochemical processes in thet pclealnl have been affected before growth inhibition occurs (Rengel, 1996). Many enzymes h baeven found to be up-regulated upon exposure to Al and these include PODs (Ezeatk ia l., 1996, Hamele t al., 1998). In transgenic Arabidopsis, expression of a POD gense idweantified to confer a degree of resistance to Al (Ezakei t al., 2000). PODs were identified to have an impor tfauntction in plant metabolism and physiology, and are conresidd eto play a role in the responses of plants to infection and abiotic stress stimuli (pGaars et al., 1985). Many plant defence responses involving PODs have been identified ahneds et include lignification (Walter, 1992), cross-linking or bonding of cell wall compnodus (Bradleye t al., 1992), suberisation or impregnation of cell walls and wound healing e(rSf het al., 1993). The gene encoding the Arabidopsis blue copper binding protein indu Acel rsesistance in yeast cells (Ezaekti al., 2000). 21 2.4.1 Physiological mechanisms of aluminium tolerance Hartwig et al. (2007) reported that Al detoxification can be accpolismhed by its complexation in the symplast with different orga nicompounds and/or by compartmentalisation of Al or its complexes in voalceus. In this case Al would change little or nothing in plant metabolism, enabling growth adnedvelopment even after Al input into the symplast. This tolerance mechanism is assodc wiaiteh endemic species of regions with acidic soils, where the ability to address Al toitxyi cis a prerequisite for survival. There are a few crop species that accumulate high conceonntrsa toi f Al in their shoots without suffering from toxicity (Ryane t al., 2001; Janseent al., 2002). According to Kochiane t al. (2004b) the main tolerance mechanisms that pro mAol te exclusion or prevent its absorption by the rooctslu idne Al immobilisation in the cell wall, Al selective permeability in the plasma membranHe ,i npcreases in the rhizosphere or the root apoplast and release of organic acids suchit raste , oxalate and malate, and phenolic compounds by the roots. The production and reloefa oserg anic acids is perhaps the major mechanism of Al tolerance. Evidence that suppohrits st tatement was discussed and concluded by Kochiane t al. (2004b) and they include: i) A strong correlation exists between Al tolerancde eaxnudation of organic acids in many crop species. ii) The addition of organic acids in the nutrient memd irueduces Al toxicity. iii) Al/organic acid complexes (di and tri-carboxylico) ndot cross the membrane and are not significantly absorbed by the roots. iv) The exudation of organic acids activated by Al orsc caut the root apex, the location of the primary effect of Al toxicity. v) The activation of the exudation mechanism is trrigegde specifically by A3l+. vi) In the plasma membrane there are anionic channcteivlsa taed by Al that facilitate the efflux of organic acids. Ma et al. (2001) also reported the identification of two pteomral patterns of organic acid exudation as follows: i) The plants are characterised by having an almomste idmiate response to the release of organic acids by the roots when exposed to Ahel. Tauthors suggested that the process 22 appears to involve the activation of pre-existinrogt epins, as found in wheat, tobacco and barley. ii) An existence of a lag-phase between Al exposure o argnadnic acids release was found and this process is assumed to involve timheu slattion of gene expression (Meat al., 2001; Magalhaeest al., 2007). It is also indicated that genotypes with more rot baunstioxidant systems are usually more tolerant to excess Al, but the mechanism by whicl ehx Aacerbates the formation of reaction oxygen species (ROS) is still not fully understo(Doda rko et al., 2004). 2.4.2 Genetic mechanism for aluminium tolerance A study using QTL mapping reported five distinctn ogme ic regions which are significant for Al tolerance in maize (Ninamango-Cárdeneat sa l., 2003). Consequently, maize has been the subject of breeding programmes seekiningc rteoa se Al tolerance or understand the basis for it. The use of nutrient solution erximpents by various authors showed that the trait is quantitatively inherited with a prevalen ocfe additive genetic effects (Lopest al., 1987; Sawasaki and Furlani, 1987). Prioli (1987g)g seusted that due to its high heritability the trait is expected to be controlled by a smuamll nber of genes. Miranda et al. (1984) found that the inheritance of two domingaennt es is responsible for tolerance to Al toxicity. Rhue t al. (1978) and, Garcia and Silva (1979) also foundt tha tolerance to Al toxicity is determined by one doamnint locus for sensitivity. In another study using restriction fragment length polymorpmh i(sRFLP) Sibove t al. (1999) also identified indicators that were involved with thwe ot loci (or two groups) located on chromosome 6 and 10. Brondani and Paiva (1996)c iastesod Al tolerance with a gene or block of genes located on chromosome 2 while To ertr easl. (1997) associated chromosome 8 to Al tolerance. Ninamango-Cárdeneat sa l. (2003) also identified five genomic regions presumably linked to maize Al tolerance, sugges tthinagt this trait is quantitatively inherited and controlled by a few genes. Their study ideendti ffiour QTLs for Al tolerance in maize located on chromosome 2, 6 and 8. Beotn ai l. (2009) suggested that different results on inheritance of Al tolerance could be a functiont hoef germplasm used because there is a possibility that the genotypic constitution of gteicn ematerial can generate a differentiated 23 and apparently inconsistent phenotypic expres sTihoen .authors recommended more efforts on the inheritance mechanisms of Al tolerance iniz em baecause the results reported in the literature are considered to be inconsistent acnodn icnlusive . 2.5 Use of modern tools in breeding for low pH toleer:a QncTLs, marker assisted selection and transgenic’s In maize, Al tolerance is seemingly a quantitattivraei t and Guimarãese t al. (2012) recommended the use of a combination of stratefogrie QsT L introgression, complemented by early phenotyping of lines to increase the cheas nocf success in generating tolerant materials. The existence of a QTL with a major cetf fceo-localised with genes homologous to the ALTSB gene was reported (Siboevt al., 1999; Ninamango-Cárdenaest al., 2003). Maron et al. (2009) recommended that other genomic regionusl ds hboe monitored on the basis of early phenotypic selection or genome-wsiedle ction (GWS). Parentoneit al. (2003) reported a high correlation between perfonrcmea per se in maize inbred lines and its general combining ability (GCA) evaluated ina ldleil crosses in a study of phenotypic selection in nutrient solution. In another studyh,r ete cycles of marker-assisted backcrossing in maize were sufficient to recovepr roaxpimately 99% of the recurrent genome (Morrise t al., 2003). Molecular markers for assisted introgression arseo alvailable for sorghum and were described by Magalhaeest al. (2007). Oliveirae t al. (2010) also reported that markers distributed in the sorghum genome, are availabdle aarne being used for introgression of superior alleles of thAe LTSB gene in elite lines from Brazil and Niger. In beayr,l transgenic plants that overexpress thAeL MT1 gene were created. Wheat genes that showed a high increase in the rate of malate release, lead tinoc arena sed tolerance to Al (Delhaiezte a l., 2004). Magalhaeest al. (2007) also demonstrated thAarta bidopsis plants transformed with the ALTSB genes showed higher Al tolerance and citrate etxiound athan non-transgenic plants. These results showed that the overexprne sosf iothese heterologous genes confers increased tolerance to Al and suggest an addit isotnratlegy for crops which have limited genetic variability for this trait. 24 Significant advances in the knowledge of the phlyosgiiocal and molecular basis for Al tolerance were obtained through the cloning of gs eonf emajor effects, such aAsM LT1 in wheat (Sasakei t al., 2004) andA LTSB in sorghum (Magalhaeest al., 2007), which are involved in the Al exclusion mechanism. Guimarãeet sa l. (2012) indicated that newly identified genes and QTL have provided importanptp sourt for a broader understanding of other mechanisms involved in Al tolerance in pla anntsd the availability of cultivars with higher levels of Al tolerance would increase in et imand efficiency with the broad integration of molecular and physiological knowled ingto breeding programmes. 2.6 Diallel evaluation The Danish animal breeder, Schmidt first coined dthiaellel crossing concept in 1919 (Pirchiner, 1979), it was later introduced inton pt labreeding. According to Sughrone and Hallauer (1997) “diallel” refers to making all poibslse crosses among a group of genotypes. The genotypes could for example, be individualosn, ecsl or homozygous lines. According to Griffing (1956) and, Mather and Jinks (1977) tdhieallel mating design enable the determination of a magnitude of additive and nodni-taivde components of heritable variation. It is the most popular technique used by plant dbererse to get information of value on inbred lines of different parents and to assthees sg ene action in various tra (itPsickett, 1993). Griffing (1956) came up with a range of diallel alyntiacal procedures. This has permitted plant breeders to come up with the right selecstitorant egies and compare heterotic patterns at an early stage of hybrid production (Go, u2i0s02). The four methods used ar pea i)rents, ii) F1 and reciprocal crosses iii) parents an1d, aFnd iv) only the 1F. Depending on the decision by the plant breeder, the linear analmysoisd el can be for either fixed or random effects. When the genotypes are highly selected i nabnrded, a fixed model for analysis is commonly engaged for applied breeding programmegsro (bAase, 2010). In this case when testing for combining ability the sampling errorc boemes the residual and consequently variance components and standard errors can bmea etesdti. When estimating additive and dominance variances, the following is assumed:n acbes oef epistasis, absence of reciprocal differences, normal diploid segregation, absenc e linokfage and multiple alleles, homozygous parents, independent gene distributaionnd, zero inbreeding coefficients 25 (Griffing, 1956). However, it has been noted thhaets te assumptions are rarely observed in practice (Baker, 1978) and since diallel cross yasnisa lguides the selection of parents with additive and non-additive effects for specific ttsr,a it enables the plant breeders to select parents to be used in hybridisation or populatiorene dbing programmes (Murtazeat al., 2005). 2.7 Combining ability analysis Griffing (1956) outlined a general procedure foar lldeil analysis which permits non-allelic interaction. This technique partitions the averamgeea sured performance of a cross into major components apart from the general mean (dµ )e annvironment varianceσ 2(e) by use of the analysis of variance (ANOVA). 2.7.1 General combining ability analysis GCA is used to denote the parents/line/hybrid’s nm peearformance/contribution in a cross combination (Sprague and Tatum, 1942). Falcone r Maancdkay (1996) defined it as the average performance of the parental inbred linael l insi ngle crosses, when expressed as a deviation from the average of all crosses. Add itaivned additive epistatic variances are the primary components of GCA (Matzinger, 1963). Advdeit,i additive x higher order interactions of additive genetic variance have b reeesnponsible for the differences in important variation in GCA (Baker, 197 8). 2.7.2 Specific combining ability anal ysis Specific combining ability (SCA) refers to thoses ecsa in which cross combinations perform relatively better or worse than would beti caipnated on the basis of the mean performance of the parental inbred lines (Spragnude T aatum, 1942). It is thus the deviation to a greater or lesser extent from the sum of tCheA Gof the two parents. SCA is due to non-additive gene action (Falconer and Mackay, 1).9 9In6 other words, variations or differences in SCA are considered to be attribuet atbol non-additive genetic variance (Baker, 1978). 26 2.7.3 Importance of combining ability analysis GCA and SCA effects are significant in identificoant iof parents and crosses which are responsible for the expression of a particular toyfp egene action (Meredith, 1984). It is important to note that both GCA and SCA are efvfeec gtienetic parameters used in deciding the next stage of the breeding programme (Dabh,o l1k9a9r 2). Multiple crossing or composite breeding programmes are facilitated tghhro suelection of parents based on GCA for development of synthetics and choice of sueit aFb1,l especially where one intends to use appropriate selection techniques like recu rsrelnetction, mass selection and reciprocal selection (Dabholkar, 1992 ). 2.7.4 Research findings on combining ability sst uind ime aize Gowda (2013) carried out a study to investigate G thCeA effects of parental inbred lines and SCA effects of single-cross hybrids for yienld ayield related traits and explore their use in the generation of hybrids. A total of 1710 w Fere developed and tested by crossing 34 parental inbred lines with five testers. The S: CGACA ratio of variances revealed that there were prevalence of non-additive gene acntio tnh ei expression of all the traits under investigation. Six inbred lines were identified hw igtood GCA for yield and yield related traits. El-Badawy (2013) carried out a study involving alf hdaiallel cross with seven inbred lines of maize under two different N levels for six quiatanttive characters. Results indicated that mean squares for all traits were significant for AG Cand SCA. Ratios of GCA:SCA indicated that the additive and additive x addi tiyvpees of gene action were responsible for the expressions for days to 50% anthesis, numb ekre ronfel rows per ear and shelling percentage in both N levels and combined analySsigisn. ificant interaction mean squares between N levels and GCA and SCA were detectedm foosrt traits. The results suggested that the crosses may be of great importance in dbinrge eprogrammes either towards development of maize hybrids or synthetic varie. ties Kurawa (2012) conducted genetic analysis studi epsro ogfenies from diallel crosses among eight varieties of different maturity groups of mzea iand observed significant differences 27 for GCA and SCA, indicating presence of additiv ew aesll as non-additive gene action. In both environments, the GCA mean squares were h isgihglnyificant and higher than the SCA mean squares for all traits with a few excenpst.i oThe study revealed significant GCA x environment interaction, indicating different epnatral varieties behaved differently under different environments, hence there was need etoc ts deilfferent parental varieties for hybrid production for a specific environment. SignificaSntC A x environment interaction indicated hybrid performance varied with respe cetn tovironments. A suggestion was made to have specific hybrids produced for specific eronnvmi ents . Vivek et al. (2009), in a combining ability study of parentnabl ried lines for grain yield and resistance to seven diseases, observed signifdicifafenrt ences for both GCA and SCA effects for most diseases. Correlations between GeCffeActs for disease scores were generally non-significant, suggesting the posstyib ioli f pyramiding genes for disease resistance in the parental inbred lines. Matthetw asl . (2008), in a study of a diallel cross among the nine lines observed that both SCA and GwCeAre significant sources of variation in the inheritance for resistance tod eaamr age. GCA was also a significant source of variation in the inheritance for resistance atorv al l growth. GCA effects for reduced larval weight were significant for two lines. 2.8 Heritability estimation Heritability was defined as the proportion of vanrciae due to heritable differences and genotypic variance to the total phenotypic varia n(Mceeredith, 1984). The higher the proportionate value, the more transmissible isc thhaer acter. On the other hand, the lower the ratio or proportionate value, the higher thfelu einnce of the environment on the phenotypic expression of the character. Thus itn dees fthe proportion of the total variance that is due to the mean effects of genes. Heritability can be defined in two sens es: i) Broad sense heritability which is total genetic iavnacre (Meredith, 1984). While Dudley and Moll (1969) defined it as the ratio oft atl genetic variance to phenotypic variance which expresses the degree to which idnudaivlsi ’ phenotypes are determined by their genotype. 28 ii) Narrow sense heritability whicihs the ratio of additive genetic variance to pheypnioct variance (Dudley and Moll, 1969) and expressese xtht en t to which phenotypes are determined by the genes transmitted from parents. 2.8.1 Importance of narrow sense haebriltity Inbreeding is important mostly when developing eindb lrines through selfing and selection, and is needed when heterosis or vigour is desired. Tvoi da mating related nidividuals (inbreeding), narrow sense heritability is emplo yine destimating the degree of relatedness or resemblance between parents and progenies (Mithe, re1d984; Chaudhary, 1991). Narrow sense heritability estimates the degreeo rorfe cspondence between breeding values and phenotypic values and expresses the leveln oeft igce variance in the population, which is mainly responsible for altering the genetic m-aukpe of the population through selection (Falconer, 1989; Dabholkar, 1992). Narrow sense heritability 2()h can be written as: 2h = VA / VP 2 h = VA / (VP = VA + VD +VI+VE) VP = VA+VD+VI +VE Where: VA denotes additive variance, D Vdenotes dominance variance,I dVenotes interaction variance, andE V denotes environmental variance (Falconer, 1989). It is important to note that heritability is valnido t only of the trait under investigation, but also of a population being sampled and the envireonntaml conditions to which individuals have been exposed to (Falconer, 1989; Dabhoka2r,) .1 P99opulations which are genetically more uniform are anticipated to express lower ahbeirliity than genetically diverse populations, since environmental variances contest iptuart of the phenotypic variance, which influences the degree of heritability. 2.8.2 Research findings on heritability studiemsa iinz e In a study to investigate broad sense heritabailnityd correlations between the traits and total grain yield, Aminu and Izge (2012) reporteedr ithability estimates for number of stands per plot, anthesis-silking interval, plaenitg ht, weight of cobs and grain yield of above 60% within a range of 60.61-67.44%, while sd taoy 50% anthesis, days to 50% silking, ear height and de-husked cobs, recorderidta hbeility estimates of below 60% i.e. 29 47.91%, 50.03%, 58.45% and 55.06% respectivelyh. eHri gand relatively moderate broad sense heritability of the traits suggested thaita vtiaorns were transmissible and had the potential for generating high yielding varietiesa vsi election of promising plants in succeeding generations. The correlation analysis which indicates assocniast ior some relationships indicated that anthesis-silking interval (r = 0.88), number of sc opber plant (r = 0.55 and r = 0.32), number of cobs per plot (r = 0.83), weight of cobs (r =8 0a.nd r = 0.86), de-husked cobs (r = 0.95, r = 0.49 and r = 0.87) and 100-seed weight (r =6 0a.n4d r = 0.32) exhibited positive and significant genotypic (g), phenotypic (p) and enovnimr ental (e) correlation with grain yield. The authors, Aminu and Izge (2012) concluded thearit ahbility as well as correlations were good methods for improving yield and selecting gtyepneos tolerant to drought. In another study to investigate genetic variatihoenr,i tability and genetic advance of grain yield and its component traits Belelot al. (2012) reported that the effect of interaction of genotype and genotype by year were significante aforr weight and grain yield, while the effect of year was highly significant for all threa its. Additive gene effects contributed to high levels of phenotypic and genotypic coefficsie onft variation as well as high heritability accompanied with high genetic advance recordegdr faoinr yield, number of grains per ear, ear weight, as well as plant and ear heights. Tinhdisic ated that effective selection is possible for improving these traits. 2.9 Heterosis Heterosis is essential in crop improvement ando inss icdered to be the phenomena of enhanced hybrid performance (Hartl and Clark, 1.9 F8a9l)coner (1989) defined heterosis as the difference between the crossbred and pal rinebnrtead lines or simply the superiority over inbred lines .I t is often expressed in two ways: Where interse sptr imarily in the F1 performancep er se, it is expressed as the1 mFinus the highest performing parent, expressed as a percentage of that parent used and is re fteor raesd “high parent heterosis” (Meredith, 1984). Secondly, as the1 Fminus mid-parent expressed as a percentage omf itdh-ep arent, this was also referred to as “mid-parent hetero (sMise”redith, 1984; Lamkey and Edwards, 1999). For heterosis to occur there should be ohzeytegrosity as a fundamental precondition 30 (Flintham et al., 1997). Other geneticists believe that dominancde eapnistasis are the essential genetic foundation of heterosis such lothcai twith no dominance do not cause heterosis. The degree of heterosis following as c broestween two particular parental inbred lines or populations is dependent on the squatrhee o df ifference of rate of gene occurrence or frequency between populations such that hetse rions tihe F1 is HF 21= ∑dy , where “d” denotes the deviation of the heterozygote from htohme ozygote mid-parent, while “y” denotes gene frequency (Coors, 1999). 2.9.1 Research findings on heterosis studies in maize Salazare t al. (1997) conducted a study to investigate eight esgeagtring populations and their 56 reciprocal crosses, in five acidic-soicl alotions in order to establish relative importance of nuclear and cytoplasmic factors fioerld y, days to silking, ear height, ears per plant, and ear rot. Average and specific hesitse rroepresented 65 and 31% of the total sum of squares for heterosis for yield. Populathioente rosis effects for yield were not significant, suggesting the effects would be otlfe l iitmportance in selecting parental inbred lines for developing superior hybrids. Specific ehreotsis effects were negative and significantly different for yield and ears per ptl aonnly for one cross, suggesting that non- additive gene effects were at play in determininiegl dy of specific cross combinations. According to the author, the unavailability of rpercoical differences for all characters suggested that nuclear genes were responsibloel eforar nt ce to soil acidit y. 2.10 Correlations When a change in one variable is associated wiathn gceh in another variable it is referred to as correlation (Falconer, 1989). The measuares soof ciation between two traits is referred to as a correlation coefficient. Correlations airthe eer positive or negative; positive when an increase in one variable leads to an increa asneo itnher or when it is negative, an increase in one variable will lead to a decrease in anot(hFearl coner, 1989). Associated traits are considered to be significant for three reasons :g e(in)e action leading into association through the pleiotropic action of genes whereby goennee affects more than one phenotypic 31 trait (ii) linked to alterations brought in by scetlieon and (iii) linked with natural selection (Falconer and Mackay, 1996). Two types of correlations are considered to be irmtapnot in plant breeding (Meredith, 1984): i) Correlation via phenotype which is considered t ot hbee association between two traits that is visible and determined from measuernetms of the same traits in a number of individuals in a given population. Phenotypic vasl uaere estimated by genotypic values and environmental deviations. This type of correlatiiosn composed of correlation due to environmental agents and non-additive gene acFtioanlc o( ner, 1989; Dabholkar, 1992). In cases where two traits have high heritability, caisastioon due to environmental agents will be insignificant (Falconer, 1989). ii) Genetic correlation is considered to be the cotriroenla of breeding values which is a result of additive gene action (Falconer and Maay,c k1996). Genetic association between two or more traits may result from one gene afnfegc tmi any traits (pleiotropic effects) or gene linkage that governs inheritance of two or em torarits (Falconer, 1989). This depicts the degree to which the two measurements are astesdo cgienetically. 2.10.1 Research findings on correlation in maize Bulent (1996) conducted a study to investigatea tshseo ciations among several agronomic characteristics in parental inbred lines, hybrindds baetween parental inbred lines and their offspring of maize in short season areas. Phenco tcyoprirelations of each measured trait between as well as among hybrids and inbred linerse wnot the same. The highest correlation coefficient (r = 0.78) was recordedw beeetn days to 50% anthesis and grain yield. The trend showed that the better yieldinbgr eind lines did not necessarily give rise to better yielding offspring. Given that days to 50o%ll epn shed plays a significant role in two main traits (yield and moisture stress tolerancoer )h fybrids, suggested that an effort to achieve most favourable anthesis dates during din blirnee development could be an important criterion to predict short season hybpreidrf ormance. Alake et al. (2008) conducted a study to determine genetica tvioanri and correlation in yield and yield associated traits of tropical maize. Teh weras close similarity between the 32 genotypic correlation coefficient and phenotypicrr ecolation coefficient for all characters, indicating that selection for these traits would sbueccessful. Heritability estimates were high for all the traits investigated except for sd atoy 50% flowering. Traits investigated showed significant association in the positive cdtiroen with grain yield except for days to flowering and silking which showed significant netivgea genotypic association with grain yield. Principal component analysis (PCA) showeda t thsome characters contributed significantly to variations found in the maize getynpoes evaluated and these included seedling emergence, kernel weight, grain yield, bneurm of kernels per row, number of kernel rows per ear and days to anthesis. Also stroamitse like number kernel rows per ear, grain yield per plant, ear length and number ofn ekelsr per row showed high heritability coupled with high genetic advance and could bei dcoenresd as criteria when selecting maize for grain yield. 2.11 Stability analysis 2.11.1 Stability definition and its concept Yield stability of a genotype refers to the situoant iwhereby a particular genotype is capable of evading fluctuations in yield over a range ovf ieronnmental conditions (Heinriceht al., 1983). Related to stability is wide adaptability icwhh refers to a situation whereby a genotype exhibits good performance over a wide rgaepohgical region under changeable climatic environmental conditions. Adaptability sotra bility of a cultivar usually relates to three biological mechanisms. These are: physioalol,g micorphological and phenological. Individual genotypes may react to brief fluctuatsi oin an environment in two different ways. Genotypes that are capable of buffering asgt aeinvironmental changes and develop a replica of phenotype over a range of environm aernet sknown to possess a ‘biological or static’ stability. This is usually not beneficianl icrop breeding, since it will not show improved vigour or heterosis under improved grow cinognditions. On the other hand ‘agronomic or dynamic’ stability allows a predicltea bresponse to environments and stability with respect to the agronomic concepdt, haans no deviation from this response to environments (Becker and Leon, 1988). With resptoe cqtu antitative characters, the many genotypes usually respond similarly to desirab luen ofar vourable environmental conditions. According to Baker (1988) genotype by environmenntet riactions (GEI) could be classified into two types: 33 i) Qualitative interactions. This is sometimes ecda llcrossover interactions and it is here that the direction of true treatment differencersie vsa. ii) Quantitative interactions. This is sometimesll ecda non-crossover interactions and the real treatment differences vary in the degree,n bout ti n direction (ranking of genotypes does not change from one environment to another). The crossover or qualitative interaction is sigcnainfit in agricultural production in comparison to non-crossover or quantitative intteiorancs (Baker, 1988; Crossa, 1990). With respect to quantitative characters, the majori tgye onfotypes respond in a similar manner to desirable or undesirable environmental conditioIfn as .c rossover (qualitative) type of GEI (one that leads to genotype rank changes) is ptr,e sseelenction of genotypes based on mean yield by use of combination of stability and yiewldo uld likely be lower than for those genotypes selected based on yield alone (Keat nagl. , 1991). This could also be clarified by investigating the outcome to growers when investotirgsa commit a Type I error i.e. rejecting the null hypothesis when it is true anydp eT II error i.e. accepting the null hypothesis when it is false, when selection wase db aosn yield alone and when it was based on both yield and stability. 2.11.2 Phenotypic stability analysis techniques Phenotype is considered to be an organism's avcitsuiballe properties, such as morphology, development, and behaviour. Several proceduresa sfsoer ssing phenotypic stability have been proposed by various authors. However, eLti na l. (1986) studied the statistical relationship between nine stability statistics aidnedn tified and classified three concepts of stability into the following types : I) Type I: a stable genotype is identified with ma asll variance across all environments. This type of stability is significant when the t elosctations considered are not very diverse and is identical to the statistical concept of islittayb as described by Becker and Leon (1988). II) Type II: a stable genotype has a reactionh eto e tnvironment similar to the mean reaction of all genotypes in the experiment. Type II statyb iilsi identical to the dynamic concept described by Becker and Leon (1988). 34 III) Type III: the residual mean square from theg rreession model on the environmental index is a small value i.e. a smaller deviationm f rothe regression. Type III stability is dynamic and the method of Eberhart and Russel ()1 i9s 6e6mployed for its estimation. 2.11.2.1 Cultivar performance technique for estiinmga ct ultivar stability Cultivar performance measurPei )( is another technique proposed by Lin and Bins8 8(1).9 Whereby P denotes a genotype whilie is the mean squares of the distance between genotype one and the genotype with the maximumo rpmerafnce. A smalPl i value indicates a small difference between the genotype with maxmim yuield and the best performing genotype. A pairwise GEI mean square between thxeim muam and each genotype can be estimated as indicated by Crossa (1990). 2.11.2.2 Wricke’s ecovalence technique for estnimg actui ltivar stability Ecovalence is another technique of Wricke (1962)o wphroposed the procedure for measuring stability by employing each genotype’nst rciboution to the GEI sum of squares as a stability measure. EcovalencWei )( was the name given to this concept or statistic. Genotypes associated wiWthi values that are low or close to zero tend to hsamvea ller deviations from the mean across environments aen dm aorre stable and possess a high ecovalence i.e. low value oWf i = high ecovalence. Becker and Leon (1988) demaotnesdt r ecovalence by plotting yield numerical values onf ogteypes in several environments against the respective environmental means. 2.11.2.3 Shukla’s stability variance parametere fsotrim ating cultivar stability Stability variance technique is another statisrtoicc pedure developed by Shukla (1972). The technique is dependent on the residuals from thdeiti vaed model. The variance of a cultivar is defined as the variance of the cultivar acroscsa tlions. For the purpose of ranking, stability variance σ(2i) is equivalent to ecovalenceW (i) (Wricke, 1962). Shukla’s (1972) stability variance σ(2i) is also equivalent to Type II stability (Lient al., 1986). According to Shukla (1972) there was an indication that gyepneost could not be described if there was 35 a small proportion of GEI due to heterogeneity agm roengression coefficients. In addition, there is absence of independence between meanitse so fa snd performance as well as between intercepts and slopes. Instead he sugg tehset eudse of the GEI sums of squares which is partitioned into variance components csoprorending to each of the genotypes. The identification of a stable genotype according tou kSlah (1972) is as follows: stability variance σ( 2i) must be equa tlo the environmental varianceσ 2(e), for a variety to be classified as stable th uσs2 = 0. A relatively large value o σf2i implies greater instability of genotypei since the stability variance is the differencew beeetn two sums of squares, the value can be negative, but negative estimates roiaf nvcaes are rare in variance components computations and are assumedv aesq uivalent to zero in most cases. 2.11.2.4 Regression coefficient and deviation msqeuaanr es Yield is considered as the most important trai tv faorriety selection. To this effect the most widely used criteria for selecting high yield antda bsle performance is average yield performance, site mean yield regression responds ed eavniations from regression (Eberhart and Russel, 1966; Freeman, 1973). The first esetidm paat rameter is the slopbei from the regression of the yields of genotype I on an ennvmireontal index (Finlay and Wilkinson, 1963). Whereb is equal or close to 1.0, it implies that a genpoet yresponds to changeable environmental conditions in a similar manner as stahme ple mean. Finlay and Wilkinson (1963) proposed that regression coefficient apphrionagc zero imply stable performance. The joint linear regression has been employedt eacsh an ique for analysing and interpreting the non-additive GEI of two way-classification d.a ta Finlay and Wilkinson (1963) proposed that regrens scioefficient close to 1.0 imply average stability. When that is accompanied witghh hmi ean yield, varieties have good general adaptation. Conversely, when accompanitehd lowwi mean yield, genotypes have poor adaptation to all test environments. Regrens vsaiolues increasing above 1.0 describe genotypes with increasing sensitivity to environmtael cnhange, for example below average stability and specifically adapted to high yield ienngvironments. Regression coefficients decreasing below 1.0 provide a measure of greauteffre rbing ability to changeable environmental conditions (above average stabilaitnyd) therefore specifically adapted to 36 low yielding environmental conditions. Stability aalynsis provides a method to establish the reaction of a genotype to changeable environtaml econnditions. 2.11.3 Multivariate techniques for stability analysis Multivariate approaches are employed in stabilintya laysis in order to obtain additional information on multivariate reaction to environmse nbty genotypes. These analyses are suited for computing two-way matrices of genotypaensd environments (Crossa, 1990). According to Becker and Leon (1988) multivariatea laynsis has three major objectives: i) to remove noise from data patterns, ii) to summea trhise data and iii) to show the structure in the data. Genotypes with similar reaction can be clustereypdo, hthesised and later evaluated and data can be easily summarised and analysed. The aimh eo fs etveral multivariate analysis techniques is to allocate test varieties into qtautaivliely homogeneous stability subsets (Becker and Leon, 1988). Within subsets, no sicgannifti GEI occurs, while differences among subsets are due to GEI. However, there amre soetbacks in multivariate analysis techniques which include: i) many dissimilarity htenciques and clustering strategies exist and selecting between them can consequently ddeipffiecrt ent cluster groups and ii) non- existent structure could find its way into the d. ata 2.11.3.1 Additive main effects and multiplicatnivter iaction analysis technique This model is observed to constitute ANOVA for gteynpoes and environment main effects with PCA of the GEI into one model with additived a mn ultiplicative parameters. It has been considered important in getting insight of pcolicmated GEI (Kang, 1996). Results are plotted in a very enlightening biplot that depibcotst h major and interaction effects for both test varieties and test locations. The additiven m eaffiects and multiplicative interaction (AMMI) model is employed to split estimated intetriaocn components and adjust mean yield for the interaction. This model partitionst ad ainto pattern rich models and eliminates noise-rich residual to increase efficiency. The nm aadivantage for using AMMI is that it gives information for a large component of variiatyb ilin its first few components with 37 successive readings indicating decreasing perce nptatgtern and decreasing percentage of noise (Purchase, 1997). The model AMMI is important in gaining insight ofo mc plicated interactions, gaining efficiency, improving selection efficiency and inecarsing experimental efficiency (Gauch, 1990). It has become an important statistical itno oidl entification of morphological and physiological characters associated with toleratnoc es tress and reveals the relative significance of several environmental factors orer ssstes (Gauch, 1993). The other merit of this model is that it can be employed in modellaingd gaining insight about interactions. According to Crossa (1990) the AMMI model is paurltaicrly useful in organising patterns and associations for genotypes and environmenets .a Tnahlysis procedure is that in the first ANOVA the total variation is partitioned into thr eoerthogonal sources, test varieties as genotypes, test locations as environments and AGsE aI. guide, Ramagosa and Fox (1993), reported that in most yield trials the proportiof ns uom of squares due to variations among environments ranges from 80-90% and that the vioanri adtue to GEI are often larger than the genotypic differences. The other uniquenesAsM oMf I analysis is that the interaction principal component analysis (IPCA) sum of squaarleosn e is often larger than that for genotype. Generally as genotypes and environmeenctosm be more diverse, GEI tends to increase and may reach 40-60% of the total vanri.a tUiosually the environmental main effect, which contributes up to 90% of the totarli avtaion, is not much relevant, especially in selection techniques. The AMMI technique produces graphs (biplots) whfiocchu s on the data structure relevant to selection, in other words on the genotype andI sGoEurces (Ramagosa and Fox, 1993). PCA partitions GEI into various orthogonal axese. rTeh is some controversy on the number of axes included in the AMMI model and how judgemtse annd indications of genetic stability can be made if too many axes are conesdid.e Gr auch and Zobel (1996) pointed out that generally AMMI 1 and AMMI 2 models with IPCA a1nd IPCA 2 respectively, are often chosen and that the graphical outlay of a exiethse, r as IPCA 1 or IPCA 2 against main effects, or IPCA 1 against IPCA 2, is not an isasnude it gives adequate information. With AMMI 3 and higher models, IPCA 3 and higher axees marostly dominated by noise, have little or no predictive value and no biological elaxnpation and can thus be ignored. Another major importance of the AMMI model is that it giv ae scost-effective means for obtaining 38 efficiency in research experiments and maximisinegn ebfits on investments (Gauch and Zobel, 1996). Gains in efficiency of yield estimoantsi have been observed to be equivalent to subsequent additional number of replication bfya cator of two to five (Crossa and Cornelius, 1993). 2.12 References Agrobase, 2010. Agronomix Software Inc., 71 Watoe rlsotr, Winnipeg, Manitoba, R3NOS4, Canad a. Ahn, S.J., M. Sivaguru, H. Osawa, G.C. Chung an dM Hat.sumoto. 2001. Aluminium inhibits the H+-ATPase activity by permanently altering the PMf ascuer potentials in squash roots. Plant Physiology 126: 1381–1390. Alake, C.O, D.K. Ojo, O.A. Oduwaye and M.A. Adeko. y2a008. Genetic Variability and Correlation Studies in Yield and Yield Related Cahcater rs of Tropical MaizeZ (ea mays L.) Journal of Agricultural Sciences, Environmeantd Technology 8:14-27. Aminu, D. and A.U. Izge. 2012. Heritability and Creolration Estimates in MaizeZ e( a mays L.) Under Drought Conditions in Northern Guinea aSnuddan Savannas of Nigeria. World Journal of Agricultural Sciences 8: 598-6 02. Aniol, A. and J.P. Gustafson. 1984. Chromosomet ilonc aof genes controlling aluminium tolerance in wheat, rye and triticale. Canadianr nJaolu of Genetics and Cytology 26: 701-705. Baker, R.J. 1978. Issues in Diallel Analysis. CSrocpie nce 18: 533-536. Baker, R.J. 1988. Tests for crossover genotype vxir oennment interactions. Canadi an Journal of Plant Scien c6e8: 405-410. Barceló, J. and C. Poschenrieder. 2002. Fast rrooowt thg responses, root exudates, and internal detoxification as clues to the mechanisomf sa luminium toxicity and resistance: a review. Environmental and Experiml eBnotatany 48: 75-92. Becker, H.C. and J. Leon. 1988. Stability analyisni sp lant breeding. Plant Breeding 101: 1-23. Bello, O.B., S.A. Ige, M.A. Azeez, M.S. Afolabi, YS. .Abdulmaliq and J. Mahamood. 2012. Heritability and Genetic Advance for Grain elYdi and its Component Characters in Maize Z(ea mays L.). International Journal of Plant Research 2: 138-145. 39 Bennet, R.R. and C.M. Breen. 1991. The aluminiumgn asli: new dimensions to mechanisms of aluminium tolerance. In: Wright, .R, B. Jaligar, V. C. and Murrmann, R.P. (Eds.) Plant-Soil Interactions at Low pH. Droercdht, Kluwer, pp.703-716. Bianchi-Hall, C.M., T.E. Carter, M.A. Bailey, M.A.R Mian, T.W. Rufty, D.A. Ashley, H.R. Boerma, C. Arellano, R.S. Hussey and W.A. oPtat.r r2000. Aluminium Tolerance Associated with Quantitative Trait Loceir Dived from Soybean PI 416937 in Hydroponics. Crop Science 40: 538-545. Boni, T.A., A.J. Prioli, S.M.A.P. Prioli, L.C. Luoci and R. de Mello. 2009. Inheritance of aluminium tolerance in maize. Crop Breeding and lAiepdp Biotechnology 9: 147- 153. Bradley, D.J., P. Kjellbom and C.J. Lamb. 1992.c iEtolir and wound induced oxidative cross linking of proline-rich plant cell wall prointe: A novel, rapid defence response. Cell 70: 21-30. Brady, N.C. 1990. The Nature and Properties of sS. Moilacmillan Publishing Company, New York, NY. Brewbaker, J.L. 1985. The tropical environment mfoari ze cultivation. In: A. Brandolini, F. Salamini (Eds.). Breeding strategies for mairzoed upction improvement in the tropics. Firenze: FAO; Instituto Agronomico per lIt’rOemare. Brondani, C. and E. Paiva. 1996 ‘’RFLP’’ analysfi sa oluminium tolerance in chromosome 2 in maize. Pesquisa Agropecuária Brasileira 315:- 579. Bulent, S., 1996. Phenotypic correlations betweeaniz em inbreds and their single cross hybrids in short season areas. Euphytica 89: 2961. -29 Calba, C.H., W.J. Horst and C. Zonkeng. 2001. M agirzaein yield correlated responses to change in acid soil characteristics after 3 yeaf rsso oil amendments’. th7 Eastern and Southern Africa Regional Maize conferencet.h -1 15th Feb. pp. 222 -227. Campbell, T.A., P.R. Jackson and Z.L. Xia. 1994f.e cEtf of aluminium stress on alfalfa root proteins. Journal of Plant Nutrition 17: 46711-4. Cançado, G.M.A., V.E.D. Rosa-Junior, J.H. Ferna,n dLe.Gz. Maron, R.A. Jorge and M. Menossi. 2005. Glutathione. S-transferase and anliumi toxicity in maize. Functional Plant Biolog y32: 1045-1055. Caniato, F.F., C.T. Guimaraes, R.E. Schaffert, VC..M A. lves, L.V. Kochian, A. Borem, P.E. Klein and J.V. Magalhaes. 2007. Genetic ditvye rfosr aluminium tolerance in sorghum. Theoretical and Applied Genetics 114: 87663.- 40 Chaudhary, R.C. 1991. Introduction to plant brege.d Oinxford and IB Publishing Co. PVT. LD. New Delhi-Bombay. Chen, J., E. Sucoff and E.J. Stadelman. 1991a. iAnlium and Temperature alteration of cell membrane permeability oQfu ercus rubra. Plant Physiology 96: 644-649. Chen, Z.F., Z.W. Zhang, Z.Y. Li and Y.F. Wang, 1b9.9 S1tudy on the correlation between the earliness of Upland cotton and its yield anbdre f iquality. China Cottons 5: 16- 17. Crossa J. 1990. Statistical analysis of multiloocna tirials. Advances in Agronomy 44: 55- 85. Coors, J.G. 1999. Selection methodology and heiste. rIons: Coors J. and Pandy S. (Ed.) Genetics and exploitation of heterosis in crops.e Aricman Society of Agronomy and Crop Science Society. Madison, pp.225-245. Crossa, J. and P.L. Cornelius, 1993. Recent demveelonpt multiplicative models for cultivar trials. In: D.R. Buxton, R. Shibles, R.A. Fortbe rBg.,L. Blad, K.H. Asay, G.M. Paulsen and R.F. Wilson (Eds.) International Crcoipe nSce. Crop Science Society of America, Madson, WI, pp. 571-577. Custódio, C.C., D.C. Bomfim, S.M. Saturnino and .N M.Bachado Neto. 2002. Estresse por alumínio e por acidez em cultivaresde soja. Scaie Angtiricola 59: 145-153. Dabholkar, A.R., 1992. Elements of Biometrical Gteicnse. Concept Publ. Camp., New Delhi, India. Darko, E., H. Ambrus, E. Stefanovits-Banyai, J. oFro, dF. Bakos and B. Barnanas. 2004. Aluminium toxicity, Al tolerance and oxidative sstrse in Al-sensitive wheat genotype and in Al-tolerant lines developed by in vitro moisccr opore selection. Plant Science 166: 583-591. Delhaize E., C.D. Craig Beaton, R.J. Bennet, V.aCg. aJdish and P.I. Randall. 1993. Aluminium tolerance in wheaTt (riticum aestivum.) Plant Physiology 103: 685-693. Delhaize, E., P.R. Ryan, D.M. Hebb, Y. Yamamoto S, Tas. aki and H. Matsumoto. 2004. Engineering high level aluminium tolerance in bya rlweith the ALMT1 gene. Proceedings of the National Academy of Scienceths eo fU SA 101: 15249-15254. Dudley, J.W. and R.H. Moll. 1969. Interpretationd aunse of estimates of heritability and genetic variance in plant breeding. Crop Scien c2e5 97:-262. Duque-Vargas, J., S. Pandey, G. Granador, G., Cloesb aHl. and E. Knapp. 1994. Inheritance of tolerance to soil acidity in tropl icmaaize. Crop Science 34: 50-54. 41 Eberhart, S.A. and W.A. Russel. 1966. Stabilitya pmaerters for comparing varieties. Crop Science 6: 36-40. El-Badawy, M.E.M. 2013. Heterosis and Combining lAityb in Maize using Diallel Crosses among Seven New Inbred Lines. Asian Journal of C Srcoipence 5: 1-13. Ezaki, B., R.C. Gardner, Y. Ezaki and H. Matsum o2t0o0. 0. Expression of aluminium- induced genes in transgenic Arabidopsis plants acmane liorate aluminium stress and/or oxidative stress. Plant Physiology 122: 66567.- Ezaki, B., S. Tsugita and H. Matsumoto. 1996. Esxspiroen of a moderately anionic peroxidase is induced by aluminium treatment ina ctocbo cells: possible involvement of peroxidase isoenzymes in aluminium ion strehssy.s iPologia Plantarum 96: 21-28. Falconer, D.S. 1989. Introduction to Quantitativen Getics, 3rd Ed. Longmans Green/John Wiley & Sons, Harlow, Essex, UK/New York. Falconer, D.S. and T.F.C. Mackay. 1996. Introdunc tioo Quantitative Genetics,t h4 Ed. Longmans Green, Harlow, Essex, UK. FAOSTAT. 2007. Food and Agriculture Organizationa tSisttical Database. FAO. FAOSTAT. 2010. Food and Agriculture Organizationa tSisttical Database: http// faostat.fao.org. Ferreira, R.P., A. Moreira and J.B. Rassini. 20T0o6x.i dez de aluminio em culturas anuais. Sao Carlos: Embrapa Pecuaria Sudeste. (Embrapaa rPiae cSuudeste. Documentos 63). pp. 35. Finlay, K.W. and G.N. Wilkinson. 1963. The analy soifs adaptation in plant breeding programme. Australian Journal of Agricultural Rersceha 14: 742-754. Flintham, J.E., W.J. Angus and M.D. Gale. 1.9 H97eterosis, over dominance for grain yield, and alpha-amylase activity in1- hFybrids between near isogenRich l dwarf and tall wheats. Journal Agriculture Scien 1c2e9: 371-378. Foy, C.D. 1988. Plant adaptation to acid aluminituomxi-c soils. Communications in Soil Science and Plant Analysis 19: 959-987. Freeman, G.H. 1973. Statistical methods for thel yasnisa of genotype x environment interactions. Heredity 31: 339-354. Furukawa, J., N. Yamaji, H. Wang, N. Mitani, Y. Mautar, K. Sato, M. Katsuhara, K. Takeda and J.F. Ma. 2007. An aluminium-activatetrda tcei transporter in barley. Plant and Cell Physiology 48:1081-1091. 42 Garcia Júnior, O. and W.J. Silva. 1979. Análiseé gtiecna da tolerância ao alumínio em milho. Ciênc. Cult. 3:1 585. Gaspar, T.H., C. Penel and H.A. Greppin. 1985. Ao- stwtep control of basic and acidic peroxidases and its significance for growth ande dloepvment. Physiologia Plantarum 64: 418-423. Gauch, H.G. 1990. Using interaction to improve dy ielstimates. In: M.S. Kang (Ed.) Genotypes x environment interaction and plant binrege. dDepartment of Agronomy. Louisiana State University, Batonrouge, Louisiapnpa., 141-150. Gauch, H.G. 1993. Prediction, parsimony and noAismee. rican Scientist 81:468-478. Gauch, H.G and R.W. Zobel. 1996. AMMI analysis ioefl dy trials. In: M.S. Kang and H.G. Gauch (Eds.) Genotype by environment interactioRnC. C Press, Boca Raton, New York, pp. 85 -122. Gouis, J. 2002. Genetic Improvement of Nutrient UEsffeiciency in Wheat, In: The Molecular and Physiological Basis of Nutrient Usfefi cEiency in Crops (eds. M. J. Hawkesford and P. Barraclough), Wiley-Blackwell,f oOrxd, UK. Gowda, R., 2013. Combining Ability Studies in Ma i(zZeea may sL.). Molecular Plant Breeding 3: 116-127. Granados, G., S. Pandey and H. Ceballos. 1993.o Rnsees pto selection for tolerance to acid soils in a tropical maize population. Crop Scie3n3ce: 936-940. Griffing, B. 1956. Concept of General and SpecCifiocm bining Ability in relation diallel crossing systems. Australian Journal of BiologSiccaile nces 9: 463-493. Guimarães, J.T.F., M.C.L. Cohen, L.C.R. Pessend.Ca,. MFrança, C.B. Smith and A.C.R. Nogueira. 2012. Mid- and late-Holocene sedimenptaroryc ess and palaeovegetation changes near the mouth of the Amazon River. Theo cHeonle 22: 359-370. Hamel, F., C. Breton and M. Houde. 1998. Isolataionnd characterization of wheat aluminium-regulated genes: possible involvemenat loufm inium as a pathogenesis response elicitor. Planta 205:531-538. Hartl, D.L. and A.G. Clark, 1989. Principles of puolaption genetics (2nd Ed.). Snauer Associates, Inc. Publishers, Sunderland. Hartwig, I., A.C. Oliveira, F.I.F. Carvalho, I. Btearn, J.A.G. Silva, D.A.M. Schnidt, I.P. Valerio, L.C. Maia, D.A.R. Fonseca and C.E.S. R 2e0is0.7. Mecanismos associados à tolerância ao alumínio em plantas. Semina: Caiêsn Acgi rárias 28: 219-228. 43 Hassan, R.M. 1998. Maize Technology Developmen tT arnadnsfer: A GIS Application for Research Planning in Kenya. CAB International. Heinrich, G.M., C.A. Francis and J.D. Eastin. 19 8S3ta.bility of grain sorghum yield components across diverse environments. Crop Sec 2ie3n: c209-212. Huang, C.F., N. Yamaji, N. Mitani, M. Yano and Y.a gNamura. 2009. A bacterial-type ABC transporter is involved in aluminium toleranince r ice. Plant Cell 21: 655-667. Jansen, S., M.R. Broadley, W. Robbrecht and E. sS.m 2e0t02. Aluminium hyper- accumulation in angiosperms: a review of its pheylnoegtic significance. The Botanical Review 68: 235-269. Jones, D.L., E.B. Blancaflor, L.V. Kochian and Sil.r oGy. 2006. Spatial coordination of aluminium uptake, production of reactive oxygenc siepse, callus production and wall rigidification in maize roots. Plant, Cell and Ernovniment 29: 1309-1318. Kang, M.S. 1996. Using genotype by environment rainctteion for crop cultivar development. Advances in Agronomy 62: 199-252. Kang, M.S., D.P. Gorman and H.N. Pham. 1991. Apaptiloicn of stability statistic to international maize yield trials. Theoretical anpdp Alied Genetics 81: 162-165. Khatiwada, S.P., D. Senadhira, A.L. Carpena, R.eSig. leZr and P.G. Fernandez. 1996. Variability and genetics of tolerance for alumini utomxicity in rice (Oryza sativa L.) Theoretical and Applied Genetics 93: 738-744. Kochian, L.V. 1995. Cellular mechanisms of alummin iutoxicity and resistance in plants. Annual Review of Plant Physiology and Plant Moleacr uBl iology 46: 237–260. Kochian, L.V., M.A. Peneros, and O.A. Hoekenga. 52.0 T0he physiology, genetics and molecular biology of plant aluminium resistance atonxdicity. Pant Soil 274: 75-195. Kochian, L.V., O.A. Hoekenga and M.A. Pineros. 2a0.0 H4ow do crop plants tolerate acid soils? Mechanisms of aluminium tolerance and phorsopuhs efficiency. Annual Review of Plant Biology 55: 459-493. Kochian, L.V., M.A. Peneros and O.A. Hoekenga. 2b0. 0T4he physiology, genetics and molecular biology of plant aluminium resistance atonxdicity. Plant and Soil 274: 175-195. Kurawa, I.A. 2012. Genetic analysis of progenieosm f rdiallel crosses among eight varieties of different maturity groups of maize. Ahmadu Be Ulloniversity, Zaria Nigeria. 44 Lagos, M.B., M.I.M. Fernandes, C.E.O. Camargo, LF.Ced. erizzi and F.I.T. Carvalho. 1991. Genetics and monosomic analysis of alumintioulmer ance in wheatT (riticum aestivum L.). Brazilian Journal of Genetics 14: 1011-1020. Lamkey, J.R. and J.W. Edwards. 1999. Quantitateivnee gtics of heterosis. In: J.G. Coors and S. Pandey (Eds.). The Genetics and Exploit aotfi oHneterosis American Society of Agronomy and Crop Science, USA, pp. 31-48. Lin, C.S. and M.R. Binns. 1988. A superiority meraes ouf cultivar performance for cultivar x location data. Canadian Journal Plant Scienc e1 9638-:198. Lin, C.S., M.R. Binns and L.P. Lefkovitch. 1986.a bSitlity analysis: Where do we stand? Crop Science 26: 894-900. Liu, J., J.V. Magalhaes, J. Shaff and L.V. Koch i2a0n0. 9. Aluminium-activated citrate and malate transporters from the MATE and ALMT fami lifeusnction independently to confer Arabidopsis aluminium tolerance. The Plaonutr nJal 57: 389-399. Liu Yong-jian, I., Duan Chuan, Tian Meng-liang, HEur- liang and Huang Yu-bi1. 2010. Yield Stability of Maize Hybrids Evaluated in Mai zReegional Trials in Southwestern China Using Nonparametric Methods. Agriculturale Sncies in China. 9: 1413-1422. Lopes, MA, R. Magnavaca A.F.C. Bahia Filho and GE..E G. ama. 1987. Avaliação de populações de milho e seus cruzamentos para tocilae raâ ntoxidez de alumínio em solução nutritiva. Pesq Agropec Bras 22:257-263. Ma, J.F. and P.R. Ryan. 2010. Understanding hownt sp lcaope with acid soils. Functional Plant Biology 37: 3-6. Ma, J.F., P.R. Ryan and E. Delhaize. 2001. Alummin ituolerance in plants and the complexing role of organic acids. Trends in Placnite Snce 6: 273-278. Ma, J.F., S. Nagao, K. Sato, H. Ito, J. Furukawda Ka.n Takeda. 2004. Molecular mapping of a gene responsible for Al-activated secretio nc iotrfate in barley. Journal of Experimental Botany 55: 1335-1341. Magalhaes, J.V., J. Liu, C.T. Guimaraes, U.G.P.a L, aVn.M.C. Alves, Y.H. Wang, , R.E. Schaffert, O.A. Hoekenga, M.A. Pineros and J.E. ffS. h2a007. A gene in the multidrug and toxic compound extrusion (MATE) faym ilconfers aluminium tolerance in sorghum. Nature Gene t3ic9s: 1156-1161. Magnavaca, R., R.C.O. Gardner and R.B. Clark. 1 I9n8h7e.ritance of aluminium tolerance in maize. In: Galbelman, H.W. and B.C. Loughmans (.E) dGenetic aspects of plant mineral nutrition. Martinus Nijhoff, Dordrecht, p 2p0. 1-211. 45 Maron, L.G., M.A. Pineros, C.T. Guimaraes, J.V. Malahgaes, J.K. Pleiman, C. Mao, J. Shaff, S.N.J. Belicuas and L.V. Kochian. 2009. Tfwunoc tionally distinct members of the MATE (multi-drug and toxic compound extruns)i ofamily of transporters potentially underlying two major aluminium toleraen cQTLs in maize. The Plant Journal 61: 728-740. Mather, K. and J.I. Jinks, 1977. Introduction too mBietrical Genetics. Chapman and Hall, London, UK. Matthews, G.A., W. P. Williams and C. A. Daves. 82.0 D0iallel Analysis of Corn Earworm (Lepidoptera: Noctuidae) Resistance in Maize. Jaolu ornf Agricultural and Urban Entomology 24: 59-66. Matzinger, D.F. 1963. Experimental estimates ofe gtiecn parameters and their applications in self-fertilizing plants. In: W.D. Hanson and H. .RFobinson (Eds.), Statistical genetics and plant breeding no. 982. NAS-NRC. Meredith, W.R. Jr. 1984. Quantitative InheritanIcne: . R.J. Kohel and C.F. Lewis (Eds.) Madison, WI .pp 131-150 Miftahudin, S.G.J. and J.P. Gustafson. 2002. AFLPar kmers tightly linked to the aluminium-tolerance gene Alt3 in ryeS e(cale cerealeL .). Theoretical and Applied Genetics 104: 626-631. Minella, E. and M.E. Sorrells. 1992. Aluminium troalence in barley: Genetic relationships among genotypes of diverse origin. Crop Scienc e5 9332-:598. Minella, E. and M.E. Sorrells. 1997. Inheritancned achromosome location oAfl p: a gene controlling aluminium tolerance in 'Dayton' barl ePyla. nt Breeding 116: 465-469. Miranda L.T., P.R. Furlani, L.E.C. Miranda and Ea.w Sazaki. 1984. Genetics of environmental resistance and super genes: Latelunmtein aium tolerance. The Maize Genetics Cooperation Newsletter 58: 46-48. Moon, D.H., L.M.M. Ottoboni, A.P. Souza, S.T. Sib, oMv., P. Gaspar and P. Arruda. 1997. Somaclonal-variation-induced aluminium-sensitive tamnut from an aluminium- inbred maize tolerant line. Plant Cell Reports 61866: -691. Morris, M.D., J.M. Ribaut and M. Khairallah. 200M3.o ney matters (II): Costs of maize inbred line conversion schemes at CIMMYT using ceonntvional and marker-assisted selection. Molecular Breeding 11: 235-247. Murtaza, N., M. Kitaoka and G.M. Ali. 2005. Gene dtiicfferentiation of cotton cultivars by polyacrylamide gel electrophoresis. Journal Ce nEtruarlope Agricultur e6: 69-76. 46 Nguyen, V.T., S. Nguyen, B.D.S. Sarkarung, C. Mnaerzti, A.H. Paterson and H.T. Nguyen. 2002. Mapping of genes controlling aluminium tonlecrea in rice: Comparison of different genetic backgrounds. Molecular Geneti6c7s: 2772-780. Ninamango-Cárdenas, F.E., C.T. Guimarães, P.R.i nMsa, rSt .N. Parentoni, N.P. Carneiro, M.A. Lopez, J.R. Moro and E. Paiva. 2003. MappinTgL Qs for aluminium tolerance in maize. Euphytica 130: 223-232. Oliveira, B.C.F.S., P.M.A. Moura, B.H. Macie, C.TG.u imaraes, F.F. Canaito, R.E. Schaffert, S.C Jr. Fonseca, S.N. Jardim-Belicuads Ja.nV. Magalhaes. 2010. Utilizacao de marcadores moleculares na selecaios tidaassc para tolerancia ao aluminio em sorgo. Sete lagos: (Embrapa Milho eg os.o Cr ircular Tecnica, 155 p7) Panda, S.K., F. Baluska and H. Matsumoto. 2009m. Ainluium stress signalling in plants. Plant signalling and behaviour 4: 592-597. Pandey, S. and C.O. Gardner. 1992. Recurrent isoenle fcotr population, variety and hybrid improvement in tropical maize. Advances in Agrono 4m8y: 1-87. Parentoni, S.N., V.M.S. Alves, A.C. Oliveira, J. gMaalhaes, A.M. Coelho, E. Schaffert, C.T. Guimaraes, E.E.G. Gama, C.L. Godoy, M.X. Ssa,n Pto.E.O. Guimaraes, C.A.P. Pacheo, W.P. Meirelles, P.H.E. Ribeiro, G.V.E. Paintad A.F.C. Bahia Filho. 2003. Genetics of aluminium tolerance in maize evaluainte dn utrient solution with and without control experiments. In: International syomsipum of plant breeding, Mexico, 2003. Book of abstracts. Mexico: CIMMYT. Parentoni, S.N., V.M.S. Alves, S.K. Milah, G.M.Aa. nCçado and A.F.C. Bahia Filho. 2001. Melhoramento para toerancia ao aluminio como fdaeto ar dapatacao a solos acidos. In: Nass L et al (Eds.) Recursos geneticos e maemlheonr to. Randonopolis, Fundacao MT, pp. 783-852. Peixoto, C.J., R.C. Alves and L.C. Cameron. 2007lu. taGmine and carbohydrate supplements reduce ammonemia increase during enncdeu rafield exercise. Universidade Federal do Estado do Rio de Janeriraoz,i lB. Peixoto, P.H.P., J. Cambraia, R. Sant’anna, P.Rs. qMuiom and M.A. Moreira. 2001. Aluminum effects on fatty acid composition and dli ppieroxidation of a purified plasma membrane fraction of root apices of two hsuomrg cultivars. Journal of Plant Nutrition 24: 1061-1070. Pickett, A.A. 1993. Hybrid Wheat – Results and Plermobs. Journal of Agricultural Science 12: 294-294. 47 Piñeros, M.A., J.E. Shaff, H.S. Manslank, V.M. Caalhrvo Alves and L.V. Kochian. 2005. Aluminum Resistance in Maize Cannot Be Solely Eixnpelda by Root Organic Acid Exudation. A Comparative Physiological Study. P lPahnyt siology 137: 231-241. Piñeros, M.A., J.V. Magalhaes, V.M. Carvalho Alvaensd L.V. Kochian. 2002. The physiology and biophysics of an aluminium tolera nmcechanism bases on root exudation in maize. Plant Physiology 129:1194-1 206. Pirchner, F. 1979. Populationsgenetik in der Ticehrzt.u 2. Aufl-, Paul Parey, Hamburg and Berlin, German. Prioli, A.J. 1987. Análise genética da tolerânc tiao xàidez ao alumínio em milhoZ e( a maysL .). Campinas, (Doctoral Thesis in Plant Genet–ic Us)n iversidade Estadual de Campinas. Purchase, J.L. 199 7P.arametric analysis to describe G x E interactinodn yaield stability in winter wheat. PhD Thesis. Department of Agronomayc,u Flty of Agriculture, University of Orange Free State, Bloemfontein, South Africpa 4. -p83 . Ramagosa, I. and P.N. Fox, 1993. Genotype x enmvieronntal interaction and adaptation. In: M.D. Hayward, N.O. Bosemark and I. Ramagosa s.()E dPlant Breeding: Principles and Prospects. Chapman and Hall, Lon pdpo.n 3,73-390. Rengel, Z. 1996. Uptake of aluminium by plant c.e Nllsew Phytologist 134: 389-406. Rengel, Z. and W.H. Zhang. 2003. Role of dynamfi cinst oracellular calcium in aluminium- toxicity syndrome. New Phytologist 159: 295-314. Rhue, R.D., C.O. Grogan, E.W. Stockmeyer and H.vLe. reEtt. 1978. Genetic control of aluminium tolerance in corn. Crop Science 18: 11006637- . Roy, A.K., A. Sharma and G. Taludker. 1988. Sompee catss of aluminium toxicity in plants. Botanical Review 54: 145-147. Ryan, P.R., E. Delhaize and D.L. Jones. 2001. Fiounn catnd mechanism of organic anion exudation from plant roots. Annual Review of PlPanhty siology and Plant Molecular Biology 52: 527–560. Ryan, P.R., J.M. DiTomaso and L.V. Kochian. 1993lu. mAinium toxicity in roots: an investigation of spatial sensitivity and the rolef tohe root cap. Journal of Experimental Botany 44: 437-446. Salazar, F. S., S. Pandey, L. Narro, J. C. Per.e zC, eHballos, S.N. Parentoni and A.F.C. Bahia-Filho. 1997. Diallel Analysis of Acid-Soil Tleorant and Intolerant Tropical Maize Populations. Crop Science 37: 1457-1462. 48 Samac, D.A. and M. Tesfaye. 2003. Plant improvem foern ttolerance to aluminium in acid soils - review. Plant Cell, Tissue and Organ Cuel t7u5r : 189-207. Sasaki, T., Y. Yamamoto, B. Ezaki, M. KatsuharaJ,. SA.hn, P.R. Ryan, E. Delhaize and H. Matsumoto. 2004. A wheat gene encoding alumin i–u mactivated malate transporter. The Plant Journal 37: 645-653. Sawasaki, E. and P.R. Furlani.1987. Genetica deara tnocl ia ao aluminio em milho cateto. Bragantia 46: 269-278. Sherf, B.A., A.M. Bajar and P.E. Kolattudkudy. 19. 9A3bolition of an inducible highly anionic peroxidase activity in transgenic tomatloa.n Pt Physiology 101: 201-208. Shukla, A.G.K. 1972. Some statistical aspects orft itpioaning genotype-environment components of variability. Heredity 29: 237-245. Sibov, S.T., M. Gaspar, M.J. Silva, L.M.M. Ottob,o nPi. Arruda and A.P. Souza. 1999. Two genes control aluminium tolerance in maize:e gtiecn and molecular mapping analyses. Genome 42: 475-482. Silva, I., T. Smyth, D. Moxley, T. Carter, N. Aleann d T. Rufty. 2000. Aluminium accumulation at nuclei of cells in the root tipsl.u oFrescence detection using lumogallion and confocal laser scanning microsc oPplya.nt Physiology. 123: 543- 552. Sivaguru, M., T. Fujiwara, J. Amaj, F. Baluka, Za. nYg, H. Osawa, T. Maeda, T., Mori, D. Volkmann and H. Matsumoto. 2000. Aluminum-induce®d3 1-b-D-glucan inhibits cell-to-cell trafficking of molecules through plaosm-desmata. A new mechanism of aluminium toxicity in plants. Plant Physiology. 1:92941-1005. Spehar C.R. 1995. Diallel analysis for mineral eelenmt absorption in tropical adapted soybeansG [ lycine max (L.) Merrill]. Theoretical and Applied Genetics0. :9707–713 Spehar C.R. and N.W. Galwey. 1996. Diallel ana lyfosri saluminium tolerance in tropical soybeans [Glycine max (L.) Merrill]. Theoreticald a Anpplied Genetics 92:267–272. Sprague, G.F. and L.A. Tatum. 1942. General VeSrspuesc ific Combining Ability in a single cross of corn. Journal of American Socief tAy goronomy 34: 923-932. Sughrone, J.R. and A.R. Hallauer. 1997. Analys isth oe fdiallel mating design for maize inbred lines. Crop Science 37: 400-405. Suhayda, C.G. and A. Haug. 1995. Citrate chelaatiso na potential mechanism against aluminium toxicity in cells: The role of calmodu.l inCanadian Journal of Biochemistry Cell Biology 63: 1167-1175. 49 Tang, Y., M.E. Sorrells, L.Y. Kochian and D.G. Gianr.v 2000. Identification of RFLP markers linked to the barley aluminium tolerancen eg Aelp. Crop Science 40: 778- 782. Taylor, G.J. 1988. The Physiology of Aluminium Troalnece in higher plants. Communications in Soil Science and Plant Analys9i:s 1 1179-1194. Torres, G.A, S.N. Parentoni, M.A. Lopes and E. Pa.a 1iv997. A search for RFLP markers to identify genes for aluminium tolerance in ma iBzera.zilian Journal of Genetics 20: 459-465. Vitorello, V.A., F.R. Capaldi and V.A. Stefanuto0. 025. Recent advances in aluminium toxicity and resistance in higher plants. Brazi liJaonurnal of Plant Physiology 17: 129-143. Vivek, B.S., O. Odongo., J. Njuguna, J. ImanywoGh.a B, igirwa, A. Diallo and K. Pixley. 2009. Diallel analysis of grain yield and resistea ntoc seven diseases of 12 African maize (Zea maysL .) inbred lines. Euphytica 172: 329-340. Von Uexküll, H.R. and E. Mutert. 1995. Global extt,e dnevelopment and economic impact of acid soils. In: R.A. Date, N.J. Grundon, G.E.y mRaet and M.E. Probert (Eds.) Plant-Soil Interactions at Low pH: Principles anda nMagement, . Dordrecht, The Netherlands, Kluwer Academic, pp. 5-19. Walter, M.H. 1992. Regulation of lignification ine fdence. In: T. Boller, F. Meins, Eds, Genes involved in Plant Defence. Springer Verlaegw, N York, pp 327-352. Wricke, G. 1962. Über eine method zur erffassungr odkeologischen Strubreite in feldversuchen. Z.Pflanzenzüchtg.47: 92-96. Wu, P., B. Zhao, J. Yan, A. Luo, Y. Wu and D. Seihnraad. 1997. Genetic control of seedling tolerance aluminium toxicity in rice. Euphytica 9278:9-293. Wu, P., C.Y. Liao, B. Hu, K.K. Yi, W.Z. Jin, J.J.i Nand C. He. 2000. QTLs and epistasis for aluminium tolerance in rice at different seendgl istages. Theoretical and Applied Genetics 100: 1295-1303. Xia, J.X., N. Yamaji, T. Kasai and J.F. Ma. 2010la.s Pma membrane-localized transporter for aluminium in rice. Proceedings of the NatioAnacla demy of Sciences of the USA 107: 18381–18385. Xue, Y.J., Tao, I. and Z.M. Yang. 2008. Aluminiunmd ui ced cell wall peroxidase activity and lignin synthesis are differentially regulateyd jbasmonate and nitric oxide. Journal of Agricultural and Food Chemistry 56: 9-6976684. 50 CHAPTER 3 A hydroponic nutrient solution experiment for ntegs ltoiw pH tolerance in tropical and sub-tropical maize genotypes 3.1 Abstract Low pH soil is a major constraint to maize prodounc tiby resource-poor farmers due to toxic levels of A3l + and P fixation. Selecting tolerant maize genot yupnedser low pH soil in the field is confounded by associated externaactl ofrs like diseases. In this study, 70 maize genotypes were evaluated in a hydroponicie nut tsrolution experiment at Chitedze Research Station in a glasshouse. The objective t ow iadsentify tolerant genotypes to the presence of KAl(SO4)2 in a nutrient solution as a proxy to low pH sooiln cditions. A completely randomised design (CRD) with four reaptleics was used. Significant differences (p<0.05) were observed for initial sneaml iroot length (ISRL), final seminal root length (FSRL), root tolerance index (Rti) annedt seminal root length (NSRL). In terms of FSRL the genotype sum of squares contributed1 %9 1o. f variation and this was supported by a heritability value of 92%. The eonnvmir ental influence contribution was low (7.6%). The NSRL, which represented the efvfeec trioot growth or elongation during the experimental period varied significantly (p<50). 0among the maize genotypes tested. The sum of squares for genotypes contributed 58to.3 t%he total variation. The heritability was 74% implying that the phenotypic differenceso nagm the maize genotypes in terms of NSRL were due to genetic differences. In terms SoRf NL, genotypes IWD C3 SYN F2-B, VPO52, and LPHpop4 were considered tolerant, an-dY DSTR SYNTHETIC-B, TZE-W POP DTC2 STR-B, TZE-YDT STR C4-B, LPHpop3, LPHpo pa1n3d LPHpop14 were sensitive to Al toxicity. The tolerant genotypese nidtified in this study will be used in the National Maize Breeding Programme as source poipounlsa to develop new low pH tolerant parental inbred lines. 3.2 Introduction Acid soils generally have low pH, contain toxic elelsv of Al and Mn and are deficient in Ca, Mg, P, potassium (K), and Mo. These charactitcesr isare known to limit the fertility of 51 acid soils (Duque-Vargaest al., 1994). Al is a major soil constituent and its tcoitxyi is observed in most acid soils where the pH level sb ealroew 5.5. Low pH is a major constraint to production of maize and other crops on tropsicoaills . Al toxicity is reported to inhibit the growth of crop plants on approximately 40% hoef tearth’s potentially arable soils (Manyowa et al., 1988). At low pH (pH<5) toxic A3l+ ions are released into the soil, and hinder root growth, this greatly affects the devpemloent of the entire plant (Kochian, 1995; Kidd and Proctor, 2000). Al toxicity causes shothrti,c k and under developed roots and plants. This significantly reduces nutrient uptakned increases susceptibility to drought (Sasakie t al., 1996). More than eight million ha of acid soilse aprlanted to maize in the tropics (Pandeye t al., 1994) and soil acidity reduces yield by about 1o0f% th e maize produced in the developing countries (Borretr oa l., 1995). Soil acidity is becoming common in most parts ofl aMwai and is limiting crop production. Continuous cultivation and burning of crop resid udeusring land preparation has contributed to the problem. More acid soils occnu trh ie high rainfall areas (>1000 mm per year) with moderate to high leaching, whereas ltkhael iane soils occur in low rainfall areas (< 500 mm per year). The bulk of very acid soiles alorcated in Lilongwe, Mzuzu and Blantyre ADDs. Chilimba (1994) reported higher Aalt usration percentages in Nkhatabay, Mulanje, Bembeke and Lunyangwa. The soils datapbraespea red by the Soils Commodity Team indicated that over 40% of the country hals p sHo iless than 5.5 (Munthali, 2007). Considerable genetic tolerance to soil acidityx ipsr essed in maize (Piñeros and Kochian, 2001). In addition, extensive genetic variabilitiyth w respect to Al tolerance exists in plants both at inter and intraspecific levels (Ishikawad aWn agatsuma, 1998). In maize, the majority of commercial genotypes are sensitive lt oto Axicity, such that breeding for more adapted cultivars seems to be the best strateigmyp troo ve farming of this crop in regions with acid soils. 3.2.1 Hydroponic nutrient solution Trejo-Téllez and Gómez-Merino (2012) defined a ineunttr solution for hydroponic systems as an aqueous solution containing mainly inorgaionnics from soluble salts of essential elements for higher plants. A hydroponic systemcu oltfi vation refers to growing plants in 52 water containing dissolved nutrients (thereforeh wouitt soil). Currently 17 elements are considered essential for most plants, these arbeo nc a(rC), hydrogen (H), oxygen 2()O, sulphur (S), copper (Cu), zinc (Zn), boron (B),o crhinle (Cl), nickel (Ni), N, K, P, Ca, Mg, Fe, Mn, and Mo (Salisbury and Ross, 1992). With etxhceeption of C and 2O, which are supplied from the atmosphere, the essential elesm aernet obtained from the growth medium. 3.2.2 Justification for use of hydroponic nutrsieonlut tion experiment Conventional breeding methods take time and arluee incf ed by environmental conditions and sometimes the expression of genes is maskedp isbtya tic interactions, whereby the expression of a particular gene depends on angoethneer. Testing of maize for Al tolerance can be effectively done in the field if the envimroennt (such as diseases and weather) can be manipulated and this could be expensive and ctiomnesuming considering the number of genotypes that need to be evaluated. Reporitcsa itned that there are three efficient and less time consuming methods used: screening inie nut tsrolutions (Magnavaceat al., 1987; Cançado et al., 1999), potted soils (Ahlrichse t al., 1990) and root staining with haematoxylin (Ruiz-Torrese t al., 1992). Among these three methods, nutrient sonlu tio screening seems to be attractive since it is lxepsesn esive and provides adequate Al stress, thereby allowing preliminary screening of a larguem nber of genotypes in a small area and consequently reduces the number of promising gepneost yto be analysed in the field (Magnavacae t al., 1987). In addition the results obtained with tshoelution culture screening method correlate positively with thostea ionbed using field screening (Gomez- Urea et al., 1996) showing that this method could be represtievnet aof what happens in the field. The effect of Al tolerance in nutrient soiolunt culture could be quantified in terms of root length measurements. Early symptoms of Al ctiotyx ioccur in the root for obvious reasons as roots are in direct contact with toxli3c+ Aions. Sensitive genotypes tend to accumulate higher amounts of Al in their root teisss u(Carvere t al., 1988). Al solution tests improve efficiency in selectioonr ftolerance because there is effective control of environmental variations, which is dcifufilt to achieve under field conditions. In the present study, KAl(S4O)2 was used as a proxy to low pH soil conditions tahned n utrient composition was according to Magnavaect aa l. (1987). The cost of liming low pH soils 53 and applying adequate P to a maize crop is beyhoen dre tach of most small-scale farmers. Maize breeding for low pH tolerance could offer oal ustion. However, conventional breeding methods take time and are influenced bvyir oennmental conditions. A more suitable solution would be to select Al toleranti zmea genotypes for use in acid soils which, in the long run, is less expensive, sustainable m aonrde environmentally friendly. The Al solution test improves efficiency of selection tfolre rance because there is effective control of environmental variations, which is difficult taoc hieve under field conditions. The objective of this study was to identify maize geynpoets that are tolerant to low pH or Al toxicity. 3.3 Materials and methods 3.3.1 Experimental materials Two hundred and ninety OPV maize genotypes wererc esodu in October 2010 from CIMMYT-Zimbabwe and CIMMYT-Colombia (267 and 23 goetnypes, respectively). The genotypes which had adequate seed for both hydirco paonnd field trials were selected. Further selections were carried out based on gteraxitnu re (a preferred trait in Malawi) and important genetic variation based on their pedisg rewehich have pre-characterised attributes. This included QPM, maize toleranced tro:u ght, DM, MSV, striga as well as yellow and orange genotypes. Finally, a total o fO 7P0Vs including those released and commonly grown in Malawi were used in the hydropco nui trient solution experiment to determine their response to Al toxicity in a hydornoipc nutrient solution experiment (Table 3.1). 54 Table 3.1 Description of the tropical and sub-tcraolp mi aize genotypes used in the study Pedigree Origin Pre-characterised traits 1 99TZEFY-STR QPM CO-B IITA QPM, STR, yellow 2 DT-WSTR SYNTHETIC-B IITA DT, STR, white 3 DT-YSTR SYNTHETIC-B IITA DT, STR, yellow 4 EVDT-Y2000 STR QPM CO-B IITA DT, STR, yellow QPM 5 EVD-W 99 STR QPM CO-B IITA QPM, STR, white 6 IAR-FLINT-Q-B IITA Flint 7 IWD C3 SYN F2-B IITA White colour 8 LOW N POOL C3-B IITA Low N 9 MULTICOB EARLY DT –B IITA DT, multiple cobbing 10 OBA SUPER1(9021-18(IITA))-B IITA QPM 11 OBATANPA/IWDC2SYNF2/IWDC2SY IITA QPM 12 OBATANPA/IWDC2SYNF2/IWDC2SYNF2-B IITA QPM 13 OBATANPA/TZLCOMP4C3F2/TZLC IITA QPM 14 OBATANPA/TZLCOMP4C3F2/TZLCOMP4C3F2-B IITA QPM 15 POP66 SR/DMR-LSRY/DMR-LSRY IITA MSV, DM 16 POP66 SR/TZUTSR-WSGY/T IITA MSV, DM, white 17 SYN DTE STR-Y-B IITA DT, MSV, yellow 18 SYN DTE STY-W-B IITA DT, MSV, white 19 TZE COPM3 DTV C2 F2-B IITA DT 20 TZE E-WPOP X LD(SET2)-B IITA White 21 TZE-W POP DTC2 STR-B IITA DT, MSV, STR 22 TZE-WDT STR QPM-CO-B IITA DT, MSV,QPM, STR 23 TZE-YDT STR C4-B IITA MSV, DT, STR 24 TZE-YPOP DTC2 STR-B IITA DT, yellow, MSV, STR 25 VPO0721 CMMYT Zimbabwe GLS, LB, MSV 26 VPO5148 CMMYT Zimbabwe GLS, LB, MSV 27 VPO5173 CMMYT Zimbabwe GLS, LB, MSV 28 VPO5179 CMMYT Zimbabwe GLS, LB, MSV 29 VPO5187 CMMYT Zimbabwe GLS, LB, MSV 30 VPO52 CMMYT Zimbabwe GLS, LB, MSV 31 VPO627 CMMYT Zimbabwe GLS, LB, MSV 32 VPO630 CMMYT Zimbabwe GLS, LB, MSV 33 VPO710 CMMYT Zimbabwe GLS, LB, MSV 34 VPO712 CMMYT Zimbabwe GLS, LB, MSV 35 VPO716 CMMYT Zimbabwe GLS, LB, MSV 36 VPO717 CMMYT Zimbabwe GLS, LB, MSV 55 37 VPO738 CMMYT Zimbabwe GLS, LB, MSV 38 VPO739 CMMYT Zimbabwe GLS, LB, MSV 39 VPO741 CMMYT Zimbabwe GLS, LB, MSV 40 VPO743 CMMYT Zimbabwe GLS, LB, MSV 41 VPO744 CMMYT Zimbabwe GLS, LB, MSV 42 VPO76 CMMYT Zimbabwe GLS, LB, MSV 43 VPO86 CMMYT Zimbabwe GLS, LB, MSV 44 VPO96 CMMYT Zimbabwe GLS, LB, MSV 45 VPO97 CMMYT Zimbabwe GLS, LB, MSV 46 LPHpop4 = Cimcali 05B ROYA1 CMMYT Colombia low pH 47 LPHpop3 = Cerrito98SCMV2B(SA7)-#-B CMMYT Colombia owl pH 48 LPHpop6 = Cimcali03HCG1A CMMYT Colombia low pH 49 LPHpop8 = GLSI01HG"A" CMMYT Colombia low pH 50 LPHpop9 = Granada 01Phaeo1AS2 CMMYT Colombia low pH 51 LPHpop10 = GRANADA01PHAE1AS1COGSCMV CMMYT Colombia low pH 52 LPHpop11 = ICAV-305 CMMYT Colombia low pH 53 LPHpop13 = PSA3 CMMYT Colombia low pH 54 LPHpop14 = S03TLYQAB05 CMMYT Colombia low pH 55 LPHpop15 = S03TLYQAB05 CMMYT Colombia low pH 56 LPHpop16 = VILLAVICENCIO01PHAEOIACLA CMMYT Colombia low pH 57 LPHpop17 = Villavicencio03Asp1C(LET-EARLY) CMMYT Clombia low pH 58 LPHpop18 = Menegua03Gloeo1C(S3) CMMYT Colombia lpoHw 59 LPHpop19 = Villavicencio03Phaeo1A CMMYT Colombia wl opH 60 LPHpop20 = MENEGUA01PHAEO CMMYT Colombia low pH 61 LPHpop21 = Caicedonia00Phaeo1A CMMYT Colombia loHw p 62 LPHpop23 = POB SIKUANI CMMYT Colombia low pH 63 LPHpop1 = Cap. Miranda 99Bact1F-1 CMMYT Colombia w lpoH 64 LPHpop2 = Cerrito98Achap2B-#-B CMMYT Colombia lowH p 65 LPHpop7 = Cimcali99BSCMVSA7Ac-#-B CMMYT Colombia wlo pH 66 LPHpop12 = menegua01cog1c(pob cog) CMMYT Colombia low pH 67 ZM309 (Msungabanja) Malawi DT, GLS MSV, LB 68 ZM523 (Mwayi OPV) Malawi DT, GLS MSV, LB 69 ZM623 Malawi GT, LB, MSV 70 ZM721 Malawi GT, LB, MSV QPM = Quality protein maize, DT = drought, MSV = Mzaei Streak Virus, LB = Leaf Blight, GLS = Gray leaf ts, po STR = Striga, SR = Streak, TZ = Tropical Zea, TZE = Triocpal Zea Early, TZL = Tropical Zea Late, IWD = Intermediate White Dent, IAR = Institute ofg Ariculture Research, COMP = composite, VPO = Populations from Dr. Viveki’s stock ID, CIMMTY-Zimbabwe, LPHpop = Low pH populations (a code to designate long pedigrees for populations) from CYIMT-MColombia, and ZM = Popular OPVs released in Mwalia, breeding stock originated from CIMMYT-Zimbabwe 56 3.3.2 Experimental procedure and design The study was carried out during the off seasoOn citno ber 2011 at the Chitedze Research Station. Hundred seeds were randomly counted oru gt epneotype and these were washed with distilled water. The seeds were then sterdil iizne 1% sodium hypochlorite solution for 5 minutes and rinsed twice with distilled water.e Tseh seeds were germinated in groups of 20 between paper-sheets moistened with distilletde rw faor 7 days, incubated at o2C7 for the first 3 days to allow good germination. Aftheer t7 days roots and shoots were visible and ready for measurement (Figure 3.1A). The ISRaLs mweasured from a random sample of 10 seedlings per genotype from the 20 seedlpinegr sp aper sheet. After measurements of the ISRL, all 20 seedlings were transplanted inlatos tpic containers (Figure 3.1B). Four of the containers per genotype which representedc raetipolni s were placed in a glasshouse and seeds were grown for 7 days in 5 L aerated nu tsrioelnuttion containing 6 ppm of Al in the form of KAl(SO4)2 (Magnavaca, 1982). The remaining container per tgyepneo was used as a control in which only the nutrient solution wasse ud i.e. no KAl(SO4)2. After 7 days from sowing the seedlings were transplanted on plasatuicz eg with a fine mesh so that only the roots of the seedlings were immersed in the nutt rsieonlution. The procedure was to open the plastic gauze with a tweezer and insert thets rouontil they are in contact with the solution. A B A: Germination in moistened paper after 7 days B: 7 days after transplanting in KAl(S4)O2 Figure 3.1 Germination of maize genotypes in nerwinsts p paper and appearance 7 days after transplanting 57 3.3.3 Nutrient solution preparation The composition of nutrient solutions used for gtrho wof maize genotypes was as described by Magnavaca (1987) (µmol eleme-nI)t: l10 900 NO3-N; 3500 Ca; 2300 K; 1300 N4H-N; 850 Mg; 590 S; 590 Cl; 25 B; 9.1 Mn; 2.29 Zn; 0.M88o ; 0.63 Cu; 77 Fe as ferric hydroxethylethylene diamine triacetate (FeHEDTAl) .w Aas added to the nutrient solution as KAl(SO4)2 and P as KH2PO4. The pH was checked and adjusted to 4.0 usinMg 0 H.1C l. 3.3.4 Data collection, measurements and calcusla otifo dnerived data After 7 days, FSRL was measured in centimetres faro rmandom sample of 10 seedlings per genotype per replication. The measurements awveerreaged and the mean ISRL and FSRL were used to calculate the relative seminoatl lreongth (RSRL) as derived data using the following equations: i) RSRL = F SRL - ISRL ISRL ii) RTi = RSRL (A+l treated plants) RSRL (A+l control plants) iii) % Response = S RL (Al+ treated plants) x 100 RSRL (A+l control plants) iv) NSRL = F SRL ISRL 58 3.3.5 Statistical analysis 3.3.5.1 Analysis of variance The data was subjected to ANOVA using Agrobase 0(2).0 A1NOVAs for each measured parameter and derived data was carried out. A g froarp NhSRL with standard errors was plotted in Excel. 3.4 Results 3.4.1 Observed symptoms of aluminium toxicity The presence of Al in the nutrient solution cauas edde lay in the vegetative growth of some maize genotypes with reduced development of nevwe lse and decreased development of shoots. Al toxicity symptoms were evident aftera 3y sd from transplanting, such as lateral root shortening, darkening and stunting. The mgaeizneo types DT-YSTR SYNTHETIC-B, TZE-W POP DTC2 STR-B, TZE-YDT STR C4-B, LPHpop3,H LpPop13, and LPHpop14 were sensitive to Al with a minimum NSRL of 0.6 cemac h. Purple colour and interveinal leaf chlorosis was observed in shoots of Al-strde smseaize plants in the susceptible genotypes after 7 days from transplanting (Figu.2re). 3 Figure 3.2 Partial view of purple colouration anhdo rstened roots observed in susceptible genotypes 59 In some susceptible genotypes purple-green codloeupri c(ting P deficiency) and chlorosis were observed. On the other hand the maize genso tIyWpeD C3 SYN F2-B (3.0 cm), VPO52 (3.5 cm) and LPHpop4 (3.0 cm), did not pret senvere symptoms of Al toxicity in their shoots (Figure 3.3). Figure 3.3 Partial view of new roots emerged frolmer at nt genotypes 7 days after transplanting 3.4.2 Analysis of variance Significant differences (p<0.05) were observedF fSorR L, RTi and NSRL (Table 3.2). In terms of FSRL the results indicated that the cbountiroi n of genotype estimated using sum of squares was high (91.1%) and this was suppobryt ead high heritability value of 96.0% (Table 3.2). The environmental influence was low.6 %(7). The repeatability, which is derived from the coefficient of determination2 )( Rwas high (92.0%). This meant that the phenotypic differences between the genotypes in trthiael were due to genotypic differences, hence little of the phenotypic diffnecres were due to environmental effects. Genotypes VPO76, TZE-WDT STR QPM-CO-B, EVD-W 99 S QTPRM CO-B, SYN DTE STY-W-B and LPHpop10 had long root lengths, valoufe s1 0.0, 9.4, 8.5, 8.5 and 8.1 cm respectively, while LPHpop7 and DT-YSTR SYNTHETIC h-Bad the shortest root length, values of 1.5 and 1.7 cm, respectively (Table A3.p2p; endix1). RTi results indicated that the influence of the egteicn component estimated using sum of squares (53.2%) was moderate and this was supp boyr tae dmoderate heritability value of 56.0% (Table 3.2). Entries SYN DTE STY-W-B, LPHpo0p, 1VPO717, VPO76 and 60 EVDT-Y2000 STR QPM CO-B had high root tolerancee ixn dvalues of 1.1, 1.0, 1.0, 0.9 and 0.8, respectively (Appendix 1). With respec tp teorcent response as described in Section 3.3.5, results indicated that the influe onfc tehe genetic component estimated using sum of squares (54.3%) was moderate and this wpapso rstued by moderate heritability value of 58.0% (Table 3.2). Entries SYN DTE STY-W, -LBPHpop10, VPO717, VPO76 and EVDT-Y2000 STR QPM CO-B had a high percent orenspe of 5.4, 4.0, 0.5, -9.0 and -15.7% respectively (Table 3.2; Appendix 1). Comepda tro the control plants, entry SYN DTE STY-W-B had the highest percent response ani odf R5t.4 and 1.1 respectively. NSRL results indicated that the influence of thnee gtiec component estimated using sum of squares (59.0%) was moderately high and this wpapso sruted by high heritability value of 74.0%. Entries VPO52, IWD C3 SYN F2-B, LPHpop4, Z0M93, had the highest NSRL of 3.5, 3.0, 3.0, and 2.5 respectively (Table 3.2)R. NL Swhich represented the effective root growth or elongation during the experimental pe rvioadried significantly (P<0.05) among the maize genotypes tested. The sum of squaregse fnoort ypes contributed 58.3% to the total variation. The heritability value from the AONVA was 74% (Table 3.2) implying that the phenotypic differences among the maize genos tyinp eterms of NSRL were due to genetic differences. On average, the roots grew c m1. 3in Al nutrient solution and the maximum was 3.5 cm. In the control, the roots g5re.6w cm on average with a maximum of 11.5 cm. Thus there was a clear effect of Ali ctiotyx on the maize genotypes (Appendix 1). VPO52 (3.5 cm) had the highest NSRL (Figure) f3o.l4lowed by IWDC3SYNF2-B and LPHpop4 with 3.0 cm each. No significant correlation was observed between NL SaRnd FSRL and Zero Al (control) treatments, as well as for percentage responseu sv eRrtsi. The rest were all significantly (P<0.01) and positively correlated except NSRL SvsR LI which were significantly (P<0.01) and negatively correlated (Table 3 .4). 61 Table 3.2 Root length measurements and derived b deaftoare and 7 days after transplanting the glassseh hoyudroponic experime nt G Pedigree ISRL FSRL Zero Al Rti % RSRL RSRL Zero NSRL Rank (cm) (cm) (cm) response Trtd (cm) Al (cm) (cm) Top 10 30 VPO52 1.37 3.45 4.37 0.79 -21.13 2.47 3.30 3.47 1 genotypes 7 IWD C3 SYN F2-B 1.40 3.46 5.07 0.68 .0-93 2 2.02 3.20 3.02 2 46 LPHpop 4 1.43 4.01 5.70 0.72 -27.84 1.99 3.31 2.99 3 67 ZM309 0.78 1.84 3.40 0.61 -38.53 1.50 3.82 2.50 4 18 SYN DTE STY-W-B 3.43 8.49 8.20 1.05 5.37 1.49 .391 2.49 5 20 TZE E-WPOP X LD(SET2)-B 1.27 2.73 3.70 0.74 .3-276 1.21 1.97 2.21 6 1 99TZEFY-STR QPM CO-B 2.85 4.99 6.73 0.74 -25.87 1.19 2.03 2.19 7 58 LPHpop 18 1.37 2.36 3.97 0.62 -38.38 1.10 2.74 2.10 8 11 OBATANPA/IWDC2SYNF2/IWDC2SY 1.40 2.89 3.97 0 .72-27.79 1.06 1.83 2.06 9 37 VPO738 1.87 3.48 4.17 0.84 -16.62 1.03 1.43 2.03 10 Bottom 10 68 ZM523 3.27 2.45 4.53 0.54 -46.28 -0.25 0.41 0.75 61 genotypes 14 OBATANPA/TZLCOMP4C3F2/TZLCOMP4C3F2-B .833 2.56 4.77 0.56 -43.75 -0.26 0.36 0.74 62 12 OBATANPA/IWDC2SYNF2/IWDC2SYNF2-B 3.93 2.81 4. 47 0.64 -36.22 -0.27 0.19 0.73 63 8 LOW N POOL C3-B 6.13 4.29 6.13 0.71 -29.39 -0 .28 0.06 0.72 64 54 LPHpop 14 6.23 4.05 6.43 0.65 -35.51 -0.36 0.03 0.64 65 23 TZE-YDT STR C4-B 5.23 3.21 4.67 0.80 -20.35 38-0 . -0.07 0.62 66 3 DT-YSTR SYNTHETIC-B 2.87 1.67 4.27 0.41 -59.28 0.3-9 0.26 0.61 67 21 TZE-W POP DTC2 STR-B 4.47 2.72 5.60 0.49 -5 0.96 -0.39 0.50 0.61 67 47 LPHpop 3 5.73 3.36 6.07 0.57 -42.74 -0.41 0.09 0.59 69 53 LPHpop 13 4.33 2.31 4.37 0.53 -46.80 -0.44 0.06 0.56 70 Mean 3.73 4.08 5.64 0.71 -29.09 0.32 0.89 1.32 LSD 0.95 1.20 1.87 0.22 22.01 0.88 1.27 0.88 Prob 0.001 0.00 0.001 0.00 0.001 0.001 0.001 0.00 CV (%) 15.8 18.3 20.5 19 46.9 171.5 88.3 42 Min 0.78 1.47 2.67 0.41 -59.28 -0.44 -0.07 0.56 Max 10.10 9.99 11.50 1.05 5.37 2.47 3.82 3.47 R-Squared 0.96 0.92 0.85 0.53 0.54 - - 0.69 Heritability 0.98 0.96 0.91 0.56 0.58 - - 0.74 MSE 0.35 0.56 1.34 0.02 185.90 0.30 0.61 0.30 LSD = Least significant difference, CV = coeffitc oief nvariation, Min = minimum, Max = maximum, MSE M =ean square error, G = genotype, ISRL = initiaml sineal root length; FSRL = final seminal root length; Zero Al = Zerou mAlinium or control; RTi = root tolerance index; RLSTRTD = relative seminal root length treated witluhm ainium; RSRLzero Al = relative seminal root length with zero alumumin;i NSRL = net seminal root length. 62 Maize genotypes Figure 3.4 Graph of nett seminal root length fonro gtyepes 63 R o o t l e n g t h ( c m ) 9 9 T Z E F Y - S T R Q P M C O - B 2 . 2 E V D T - Y 2 0 0 0 S T R Q P M C O - B 1 . 2 E V D - W 9 9 S T R Q P M C O - B 1 . 2 I A R - F L I N T - Q - B 1 I W D C 3 S Y N F 2 - B 3 L P H p o p 1 0 . 9 L P H p o p 1 0 1 . 6 L P H p o p 1 1 1 L P H p o p 1 2 0 . 9 L P H p o p 1 5 1 . 2 L P H p o p 1 6 1 . 2 L P H p o p 1 7 1 L P H p o p 1 8 2 . 1 L P H p o p 1 9 1 L P H p o p 2 1 . 5 L P H p o p 2 0 1 . 7 L P H p o p 2 1 1 . 4 L P H p o p 2 3 1 . 2 L P H p o p 4 3 L P H p o p 6 1 . 7 L P H p o p 8 1 . 2 L P H p o p 9 0 . 9 O B A T A N P A / I W D C 2 S Y N F 2 / I W D C … 2 . 1 P O P 6 6 S R / T Z U T S R - W S G Y / T 1 . 5 S Y N D T E S T R - Y - B 1 . 6 S Y N D T E S T Y - W - B 2 . 5 T Z E C O P M 3 D T V C 2 F 2 - B 1 . 3 T Z E E - W P O P X L D ( S E T 2 ) - B 2 . 2 T Z E - W D T S T R Q P M - C O - B 1 . 1 T Z E - Y P O P D T C 2 S T R - B 1 . 4 V P O 0 7 2 1 1 . 3 V P O 5 1 4 8 1 . 5 V P O 5 1 7 3 1 . 6 V P O 5 1 8 7 1 . 4 V P O 5 2 3 . 5 V P O 6 2 7 1 . 9 V P O 6 3 0 1 . 7 V P O 7 1 2 1 . 2 V P O 7 1 6 1 . 3 V P O 7 1 7 1 . 7 V P O 7 3 8 2 V P O 7 4 1 1 V P O 7 4 3 1 . 2 V P O 7 4 4 1 V P O 7 6 1 V P O 8 6 1 . 4 V P O 9 6 1 . 5 V P O 9 7 1 . 4 Z M 3 0 9 2 . 5 Z M 6 2 3 1 . 2 Z M 7 2 1 1 . 3 Table 3.3 Genetic and phenotypic variances andta hbeilriity estimates from ANOVA for the measured danerdi ved data Variable Genetic variance Phenotypic variance aBd-rsoense heritability R-Squared CV (%) P for entry 1 ISRL 4.78 5.18 0.90 0.95 17.2 *** 2 FSRL 4.28 4.80 0.89 0.92 18.3 *** 3 NSRL 0.27 0.57 0.47 0.69 41.5 *** 4 Rti 0.0086 0.027 0.32 0.54 19.2 *** 5 % Response 85.68 271.595 0.32 0.54 46.9 *** 6 Al Control 4.40 5.78 0.76 0.85 20.6 *** CV = coefficient of variation, ISRL= Initial semli nroaot length; FSRL = Final seminal Root lengthR; NL S= Net seminal root length; RTi = Root toleranincdee x; Al Control = Aluminium control, *** P for entry fmro ANOVA Table 3.4 Pearson’s coefficient of correlation agm thoen measured and derived data Variable ISRL FSRL Zero Al(control) Rti % response FSRL 0.72** Zero Al(control) O.76** 0.91** Rti 0.22** 0.56** 0.19** %Response 0.22** 0.56 0.19** 1.00 NSRL -0.49** 0.08 -0.03 0.27** 0.27** **P<0.01; ISRL = initial seminal root length; FSR=L f inal seminal root length; NSRL = net seminatl rleonogth; RTi = root tolerance index; Al controall u=m inium control 64 3.5 Discussion The results indicated that after 7 days from tralannstping into the hydroponic nutrient solution, there was a general reduction in roowt gthro of all the tested genotypes at low pH of 4.0 and KAl(SO4)2. The change in colour from green to purple in esupsticble genotypes could be attributed to low P stress as it is fixuendd er low pH conditions. Purple discolouration is among the key symptoms for Ps st rien maize plants. Al toxicity is reported to be responsible for significant chaning ebsio chemical and structural patterns of plant cells reflecting on reduction of cell multicipaltion (Minochae t al., 1992) and cell growth altering auxin action in cell wall (Meat al., 1999). The root is the plant organ most affected by Al toxicity and more specifically, ittisp is considered to be the main site for Al toxicity (Archambaulte t al., 1997). As a result, root elongation is consid etore be the most sensitive parameter in a short period of time ahnuds tmay represent the whole plant reaction to Al. Noblee t al. (1988) demonstrated that it was possible to obes edravmage within 24 hr caused by Al in roots of soybean psla ndtirectly proportional to Al concentration in the nutrient solution. In susceptible genotypes the common symptom is polaonrt growth as a result of root injury (Meda and Furlani, 2005; Juan-Pientg a l., 2006). This is because root development plays a major role in a plant’s response to wantedr n autrient availability. Poorly developed roots negatively affect exploration of bulk soidl urecing nutrient and water uptake (Okiyo et al., 2010). The observation of Al toxicity symptomlisk,e root darkening and root-tip stunting in this hydroponic experiment could bea tredl to the rapid entry of Al into the root cells. In the susceptible genotypes the rowoetrse thick and shortened and this is in agreement with Blancafloert al. (1998) and Horset t al. (1999). They reported that the root morphological alterations like shrinking and tipr lcinug, observed in roots of some of the plant materials, could be attributed to alterat ioinn sthe root’s cytoskeleton. The authors also demonstrated that changes in the organisatniodn stability of microtubules and microfilaments in root cells of maize were corredla tto Al toxicity, besides the rapid inhibition of root elongation. Braccineit al. (2000) also reported that in the presence of Al, root elongation of coffee genotypes were more ateffde cthan root dry matter. 65 In the present study, root elongation was the pbaersat meter to make comparisons among maize genotypes with respect to Al tolerance. sIot aalppeared to be a cheaper and better technique compared to other methods reported olioket s rtaining with haematoxylin (Ruiz- Torres et al., 1992). According to Concado et al (1999) rootc easp i are excised and photographed both stereoscopic and light microssc opaefter staining with 0.2% haematoxylin (Merck) and to observe the presenc hea oefmatoxylin – Al complexes in internal tissues you need to carryout transverescatilo sning of the apices. This suggest more labour costs and time for this method considerhineg l at rge number of genotypes evaluated in breeding programmes. The root is the first p loarngtan to be affected as it is directly in contact with the toxic A3l+ ions. The shoot can be difficult to measure ams aity require processing dry weight and fresh weight biomassg u dseinstructive sampling. Reports also indicate that shoots do not provide a true picotunr eA l tolerance (Mascarenheats al, 1984; Bernai and Clark, 1998). Bernai and Clark (1998s)e orvbed that differences in dry matter production in sorghum genotypes were not as sicgannifti as root elongation, the latter being the best discriminative parameter. Root elongahtiaosn been considered the most sensitive characteristic to quantify Al tolerance due to ftahcet that the elongation zone is the site where Al toxicity is primarily detected (Blancaf leotr al., 1998). The most sensitive site for Al action in the root is the distal transition zo (nDeTZ) (Kollmeier et al., 2000). These authors showed that application of Al solution euxscivlely in the DTZ inhibited root elongation in a similar pattern to whole-root acpaptlion. The differential tolerance to Al by the genotypes tested could be related to Alu esxiocnl mechanisms (Silveat al., 2000; 2002) and/or symplast tolerance (Wataneatb ael ., 2001). Other research findings indicated that exclusion mechanisms are based on the rednu ocft iAol3+ activity in root tips, like the exudation of low molecular weight organic compou, nwdhsich may form stable complexes with Al, reducing its toxicity to roots, such ast racite (Miyasakae t al., 1991), malate (Delhaizee t al., 1993), polypeptides (Baseut al., 1994) and flavonoids (Kidedt al., 2001). More than one type of organic acid may be releabsye Ad l stressed roots (Larseent al., 1998; Mae t al., 2000). 3.6 Conclusions and recommendations It was possible to demonstrate that nutrient sonlu tisi a practical and efficient tool for evaluating maize genotypes for low pH toleranceo. tR eolongation appears to be the best 66 parameter to compare for Al tolerance among geneost.y pIn this experiment genotypes IWD C3 SYN F2-B, VPO52 and LPHpop4 were considehreigdh ly tolerant, SYN DTE STY-W-B, ZM309 were considered tolerant and 99TZE-SFTYR QPM CO-B and OBATANPA/IWDC2SYNF2/IWDC2SY moderately tolerant a nd DT-YSTR SYNTHETIC-B, TZE-W POP DTC2 STR-B, TZE-YDT STR C4,- BLPHpop3, LPHpop13, and LPHpop14 were sensitive to Al toxyi.c Citonventional breeding methods will continue to be used until Al-tolerant geneos mfr various plants and/or microbes are isolated and transgenic plants with significanntlcyr ei ased Al tolerance are produced. The highly tolerant genotypes identified in this stundaym ely IWD C3 SYN F2-B, VPO52 and LPHpop4 will be used in the National Maize Breed Pinrgogramme in Malawi as source populations to develop new low pH tolerant pare nintablred lines using the pedigree method. 3.7 References Agrobase. 2010. Agronomix Generation II Softwanrec,. I71 Waterloo St. Winnipeg, Manitoba, R3N054, Canad a. Ahlrichs J.L., M.C., Karr, V.C. Baliggar and R.J. riWght. 1990. Rapid bioassay of aluminium toxicity in soil. Plant and Soil 122: 2-72985. Archambault, D. J., G. Zhang and G. Taylor. 199p7a. tSial variation in the kinetics of Aluminium (Al) uptake in roots of wheatT r(iticum aestivum L.) exhibiting differential resistance to Al. Evidence for metaisbmol-dependent exclusion of Al. Journal of Plant Physiology 151: 668-674. Basu, U., A.G. Good and G.J. Taylor. 2001. Traniscg eBnrassica napusm anganese superoxide dismutase cDNA are resistant to alumin. iPulant, Cell and Environment 24: 1269-1278. Basu, U., A. Basu and G. J. Taylor. 1994. Diffetirael nexudation of polypeptides by roots of aluminium resistant and aluminium-sensitive icvualrts of Triticum aestivumL . in response to aluminium stress. Plant Physiology 15016-:158. Bernai, J.H. and R.B. Clark.1998. Growth traits anmg osorghum genotypes in response to aluminium. Journal of Pant Nutrition 21: 297-305. 67 Blancaflor, E. B., D. L. Jones and S. Gilroy. 19 A98lt.erations in cytoskeleton accompany aluminium induced growth inhibition and morpholoagli changes in primary roots. Plant Physiology 118: 159-172. Borrero, J.C., S. Pandey, H. Ceballos, R. Magna vaancda A.F.C. Bahia-Filho. 1995. Genetic variances for tolerance to soil aciditay itnro pical maize production. Maydica 40: 283-288. Braccini, M. C. L., H. E. P. Martinez, A. L. Bracnci iand S. M. Mendonça. 2000 Avaliação do pH da rizosfera de genótipos de café em res pào tsotxaidez de alumínio no solo. Bragantia 59: 83-8. 8 Cançado, G.M.A., L.L. Loguercia, P.R. Martins, S P.Nre. ntoni, E. Paiva, A. Borem, M.A. Lopes.1999. Heamatoxylin staining as a phenotynpdice xi for Al tolerance selection in tropical maize Z(. mays L.). Theoretical and Applied Genetics 99: 747-7 54. Carver, B.F., W.P. Inskeep, N.P. Wilson and R.L.s Wtmean. 1988. Seedling tolerance to Aluminium toxicity in hard red winter wheat germspmla. Crop Science 28: 463-467. Chilimba, A.D.C. 1994. Annual Report: Soil Fertyil itand Plant Nutrition Commodity Team, Department of Agricultural Research Serv icMesin,istry of Agriculture, Lilongwe, Malawi. Delhaize, E., P. R. Ryan and P. J. Randall. 199lu3m. Ainum tolerance in wheaTt r(iticum aestivumL .).II. Aluminum-stimulated excretion of malic ac firdom root apices. Plant Physiology 103: 695-702. Duque-Vargas, J., S. Pandey, G. Granador, H. Coesb anlld E. Knapp. 1994. Inheritance of tolerance to soil acidity in tropical maize. Cp rSocience 34: 50-54. Gomez-Urea, R., C. Hernan, P. Shiraji, F.C. Ant,o nFi.o Bahia and A.L. Lous. 1996. A greenhouse screening technique for acid tolerann cmea iize. Agronomy Journal 88: 806-812. Horst, W.J., N. Schmohl, M. Kollmeier, F. Baluskan, d M. Sivaguru. 1999. Does aluminium affect root growth of maize through inatcetrion with the cell wall-plasma membrane cytoskeleton continuum? Plant and Soi:l 126135-174. Ishikawa, S. and T. Wagatsuma. 1998. Plasma meme bprearnmeability of root tip cells following temporary exposure to Al ions is a rapmide asure of aluminium tolerance among plant species. Plant Cell Physiology 39: 52156.- 68 Juan-Ping, W., H. Raman, G.M.N. Zhang and M.X. Z. h2u0o06. A3l + tolerance in barley (Hordeum vulgar eL) Physiological mechanisms, genetics and screge mninethods. Journal of Zhejing University 7: 769-787. Kidd, P.S. and J. Proctor. 2000. Effect of Alummini uon the growth and mineral composition ofB etula pendul aRoth. Journal of Experimental Botany 51: 1057-1. 066 Kidd, P.S., M. Llugany, C. Poschenrieder, B. Guannsde J. Barcelo. 2001. The role of root exudates in aluminium resistance and silicon–ind uacme elioration of aluminium toxicity in three varieties of maizeZ e(a may sL.). Journal of Experimental Botany 52: 1339-1352. Kochian, L.V. 1995. Cellular mechanisms of Alumin utomxicity and resistance in plants. Annual Review of Plant Physiology and Plant Moleacr uBl iology 46: 237–260. Kollmeier, M., H.H. Felle and W.J. Horst. 2000. Goetynpical differences in aluminium resistance of maize are expressed in the distta lo pf athre transition zone. Is reduced basipetal auxin flow involved in inhibition of ro oetlongation by aluminium? Plant Physiology 122: 954-956. Larsen, P. B., J. Degenhardt, C.Y. Tai, L.M. Stern,z Sl . H. Howell and L.V. Kochian. 1998. Aluminum resistant Arabidopsis mutants thxaht ibeit altered patterns of aluminium accumulation and organic acid releasme froots. Plant Physiology 117: 9-18. Ma, J. F., R. Yamamoto, D. Nevins, J. H. Matsumaontod P. H. Brown, 1999. Al binding in the epidermis cell wall inhibits cell elongat ioonf okra hypocotyl. Plant Cell Physiology 40: 549-556. Ma, J. F., S. Taketa and Z. M. Yang. 2000. Alumin tuomlerance genes on the short arm of chromosome 3R are linked to organic acid relea streit iicnale. Plant Physiology 122: 687-694. Magnavaca, R., R.C.O. Gardner and R.B. Clark. 1 I9n8h7e.ritance of aluminium tolerance in maize. In: Galbelman H.W. and B.C. Loughman (.E) dGsenetic aspects of plant mineral nutrition. Martinus Nijhoff, Dordrecht, p 2p0. 1-211. Magnavaca, R. 1982. Genetic variability and thee rinitahnce of aluminium tolerance in maize (Zea may sL.). Lincoln, University of Nebraska. PhD thesis. Manyowa, N.M., T.E. Miller and B.P. Foster. 1988.li eAn species as sources for Aluminium tolerance genes for wheTart iticum aestivum. Proceedings of the 7th International Wheat Genetics Symposium pp. 851- 857. 69 Mascarenhas, H. A. A., Camargo, C. E.O. and Faeli,v Sen. M. P. 1984. Efeito do alumínio sobre o crescimento de raízes, peso seco da péarretea ae raízes de diferentes cultivares de sojaB. ragantia, 43, 191-200. Meda, AR. and P.R. Furlani. 2005. Tolerance to A tlo3x+icity by tropical Leguminous plants as cover crops. Brazilian Archives of Bioyl oagnd Technology. 48: 28-32. Minocha, R., S.C. Minocha, S.L. Long and W.C. Slheo. r1t992. Effects of aluminium on DNA synthesis, cellular polyamines, polyamine binotshyetic enzymes and inorganic ions in cell suspension cultures of a woody plaCnat,t haranthus roseu. sPlant Physiology 85: 417-424. Miyasaka, S.C., J.G. Buta, R.K. Howell and C.D. .F 1o9y91. Mechanisms of aluminium tolerance in snap beans. Root exudation of citcriidc. aPlant Physiology 96: 737-743 Munthali, M.W. 2007. Integrated Soil Fertility Magneament Technologies: A Counteract to Existing Milestone in Obtaining Achievable Ecomniocal Crop Yields in Cultivated Lands of Poor Smallholder Farmers in Malawi. In: vAandces in intergrated soil Fertility Management in Sub-Saharan Africa: Chagllesn and opportunities, pp. 531- 536. Noble, A.D., M.V. Fey and M.E. Sumner. 1988. Camlc-iualuminium balance and the growth of soybean roots in nutrient solutions. SSocilience Society of America Journal: 52: 1651-1656. Okiyo, T., S. Gudu, O. Kiplagat and J. Owuoche,0 2. 0C1ombining drought and aluminium toxicity tolerance to improve sorghum productivAitfyr ican Crop Science Journal 18: 147-154. Pandey, S., H. Ceballos, R. Magnavaca, A.F.C. B, aJh. iDauque-Vargas and L.E. Vinasco. 1994. Genetics of tolerance to soil acidity in ticroapl maize. Crop Science 34: 1511- 1514. Piñeros M.A. and L.V. Kochian. 2001. A patch-clamstpu dy on the physiology of Al toxicity and Al tolerance in maize. Identificatioann d characterization of Al 3+ - induced anion channels. Plant Physiology 125: 2 39025 –. Ruiz-Torres, N.A., B.F. Carver and R.L. Westman9. 21.9 Agronomic Performance in acid soils of wheat lines selected for haematoxylinn sintagi pattern. Crop Science 32: 104- 107. Salisbury, F.B. and C.W. Ross. 1992. Plant Phygsiyo. loWadsworth Publishing Company, ISBN 0-534-1562-0 California, USA. 70 Sasaki, M., Y. Yamamoto and H. Matsumoto. 1996. niLnig deposition induced by aluminium in wheat T(riticum aestivum) roots. Physiologia Plantarum 96: 193–1 98. Silva, I.R., T.J. Smyth, D.F. Moxley, T.E. CarteNr.,S . Allen and T.W. Rufty. 2000. Aluminum accumulation at nuclei of cells in the tr otiop. Fluorescence detection using lumogallion and confocal laser scanning msicoropy. Plant Physiology 12: 543-552. Silva, I.R., T.J. Smyth, N F. Barros and R.F. Nosv. a2i002. Physiological aspects of aluminium toxicity and tolerance in plants. In: Aarlvez, V. H., Schaefer, C. E. G. R., Barros, N. F., Mello, J. W. V. and Costa, L. M. s(E.)d Tópicos em ciência do solo. Viçosa: Sociedade Brasileira de Ciência do sol o2.7 27:-335. Trejo-Téllez L.I. and F.C. Gomez-Merino. 2012. Nieuntrt Solutions for Hydroponic systems, Hydroponics - A Standard Methodology floarn tP Biological Researches. Colegio de Postgraduados, Montecillo, State of Mcoe.x Mi exico. Watanabe, T., M. Osaki and T. Tadano. 2001. Al kuep tkainetics in roots oMf elastoma malabathricumL . - an Al accumulator plant. Plant and Soil 23813: -2291. 71 CHAPTER 4 Phenotypic evaluation for tolerance to low pHo inp itcral and sub-tropical maize germplasm 4.1 Abstract Low-soil pH is one of the major abiotic factors ctroibnuting to low yields in Malawi. In this study 45 maize genotypes were evaluated uas (in0g, 1) alpha lattice design with three replications, using optimal and low pH sites foro twconsecutive seasons. The objective was to evaluate genetically diverse maize geno tfyopre tsolerance to low pH soils by using phenotypic traits. A mean reduction in grain yiaenldd yield components (100 seed weight, plant vigour etc.) of 69.9% was recorded under ploHw c onditions. Genotypes LPHpop16, LPHpop3, VPO739, VPO5173 and LOW N POOL C3-B pemrfeodr relatively better under low pH conditions. Comparison between the glasseh ohuysdroponic trial and low pH field trial, indicated that among the top 10 yielding ogteynpes, SYN DTE –STY-W-B performed well and ranked first in terms of root tolerancdee inx (RTi) with a net seminal root length (NSRL) of 2.5 cm followed by VPO717 with RTi of 1 .c0m and NSRL of 1.7 cm. Phenotypic traits associated with grain yield, s uacsh plant vigour, 100 seed weight, shelling percentage, number of ears per plant,h eaigrh t and plant height can be used alongside grain yield when selecting germplasmt ofolerr ance to low pH stress. Clustering of the genotypes based on morphological data a clorows spH environments showed that genotypes had similar chances of being groupedn yin oaf four clusters. The five top yielding genotypes across low pH and optimal ennvmiroents were comprised of two genotypes from CIMMYT-Colombia and three from CIMMT-YZimbabwe. Under low pH conditions the top five yielding genotypes consdi sotfe four from CIMMYT-Zimbabwe and one from CIMMYT-Colombia. Selection for low pH toralence could be effectively carried out in situ. 4.2 Introduction Malawi is located to the south east of Africa asn da iland locked country, extending from 9◦45’ to 17◦5’ south of the equator and is about 900 km in tlhe nfrgom north to south and 72 varying in width from 80 to 160 km. Its populatioisn e stimated at about 14 million people and is increasing with an average growth rate eastetidm at 3.3% per year (Saekta a l., 2004). Malawi is endowed with many natural resources.m Itasi n resource base is agriculture, which plays a key role in the country’s economye. Tahgriculture sector remains the engine of economic growth in Malawi as it contributes o v4e0r% to GDP of which over 70% is generated by the smallholder sector; 90% of exepaorrnt ings, provides employment to over 85% of the country’s population and is a sourcein coof me for over 60% of the rural poor (Kumwenda and Kachu,l e2003). Maize is the most important staple food in Malauwci hs that in certain districts other foods like rice, sweet potato and cassava are considaesr esdn acks and households will look for resources to buy maize. Approximately 70% of thletiv cauted area of the customary land is planted to maize each year. About 1.2 million h la nodf in the country are planted to maize, producing an average grain yield of 1.1 MT-1 h(aMOA, 2003; 2005) against a potential of 6 MT ha-1 and 8 MT ha-1 for OPVs and hybrid maize, respectively (Zambeet zai l., 1993). There clearly exists a yield gap between potenytiael ds achievable and what the smallholder farmers are currently producing. Thea llcehnge for the country to attain sustainable maize productivity growth is to redtuhcee existing yield gap. The introduction of targeted input subsidy in 20im07p roved national yield performance, but still at household level some yield gaps eaxgisati nst potential yields achievable due to inadequate availability of stress tolerant varise tfioer some ecologies. FAOSTAT (2013) reported that performance of maize in Malawi remeda in early static through the mid- 2000s but it seems to have picked up in recents y. eFaarrmers are being encouraged to combine proper use of subsidised fertiliser and pcoosmt manure to improve the maize yields. Unfortunately the quality of compost man uisre compromised by the organic materials used due to poverty levels. Reducing ytiheeld gap at household level and obtaining potential yields achievable requires cionmedb efforts. These include breeding for stress tolerance in maize and use of effectiven aogmroic practices, bearing in mind that the country’s population is increasing while land iast isct. Malawi upland soils are highly weathered and low p Hin and available P. About 40% of the country’s soils are oxisols and ultisols (USxo tnaomy), which are low in pH and P 73 (Munthali, 2007). Apart from low pH, low N soils ea ralso a big problem in highly populated areas. Crop rotation with fallow is rya rperlacticed. Most farmers are practicing mono-cropping and in some places it is associatiethd rwemoval of crop residues after harvest which could be depleting soil nutrients oarngdanic matter. This has led to low-soil fertility, especially in the mid-altitude ecolog6y0 [0-1300 metre above sea level (masl)] of the country. 4.3 Materials and methods 4.3.1 Description of sites The study was conducted at Research station sroitmes 2f011 to 2013. The sites included four low pH sites and five optimal sites. In 201h3e tgenotypes were also evaluated at a managed low N site. The sites were Tsangano Expeenrtiaml Research Station (1524 masl, 1071 mm annual rainfall, sandy clay loam soil, aanl nmuaximum (max) temperature of 22.9oC). Bembeke Research Station - a low pH site (1m5a2s4l , 1053 mm annual rainfall, sandy clay loam soil, 23o.C4 annual max temperature). Bvumbwe Research S t-a ati olonw pH site (1174 to 1228 masl, 1219 mm annual rai n2f5a.ll2oC annual max temperature), Lunyangwa Research Station - a low pH site (134s2l ,m 1a270 mm annual rainfall, 3oC2 annual max temperatures). Baka Research Statiwon-l,a lnod (786 masl, 800 mm annual rainfall, sandy loam soil, 28o.C4 annual max temperatures). Chitedze Researcho nS t-a ti low N managed site (1219 masl, 954 mm annual rlal,i nsfandy clay loam, 26o.2C annual max temperature). Chitala Research Station, lowd- (la6n10 masl, 800 mm annual rainfall, 28.3oC max annual temperature). 4.3.2 Experimental materials A total of 202 genotypes were obtained from CIMMYCTo-lombia and CIMMYT- Zimbabwe. These were used in preliminary screeninin fgo ur sets of field trials in the 2010/11 season to come up with a workable set noof tgyepes (results not shown). In the 2011/12 and 2012/13 seasons a total of 45 geno twyeprees planted at 10 sites. Data from Bvumbwe sites for 2011/12 and 2012/13 was not dinecdlu in the combined analyses due to gaps in the trial for 2011/12 as a result of curt mw oand wire worm damage. The managed 74 low N site data for 2011/12 at Chitedze was alsto i ncoluded due to vandalism before harvest but the results for the 2012/13 site wepreo rrted separately. The genotypes were selected from the list provided in Chapter 3 (T a3b.1le). 4.3.3 Experimental design The experimental design used was a (0, 1) alphtiace l awt ith three replications. Plot sizes were: two rows of 5.1 m per plot with 17 plantintagt isons per row. Two seeds were planted per station and thinned to one while leaving tw ob oatth ends of the row. The phenotypic and agronomic traits measured in the study ared l iisnt Table 4.1. 4.3.4 Salient management activities 4.3.4.1 Fertilizer application The Malawi government recommended rate of 92 kgn dN 6 a0 kg P2O5 was applied as split application as follows: 131 g of 23:21:0 + 4S wpapsl iaed per row or 8 g per station if using the dollop method as basal dressing. This was eadp pwlithin 14 days after emergence. This was followed by an application of 54 g of urea rpoewr or 3 g per station if using the dollop method. This top dressing fertilizer was applied d 2a1ys from planting as recommended (MOA, 1994). 4.3.5 Soil characterisation for low pH sites 4.3.5.1 Soil sampling and laboratory analysis Soil samples were collected from four low pH sinte 2s0 10 using the grid sampling method at 15 cm depth as top soil and at 30 cm as su.b -Fsioveil samples of the same depth were thoroughly mixed to form a composite sample fromic wh hthree sub-samples were prepared for analyses. The samples were analysed at thes Saonidl Agriculture Engineering Laboratory at Chitedze Research Station. Param eatnearlsysed included pH, % organic carbon, % organic matter, % N, P, K, Ca, Mg and Al. 75 Table 4.1 List of phenotypic and agronomic tranitds maeasuring procedure No Abbreviation Trait Units Trait description 1 AD Days to anthesis Days The number of days fprolamn ting to 50% pollen shedding 2 DS Days to silking Days The number of days frolamn tping to 50% silking 3 PH Plant height cm Measured from the ground scuer tfoa the flag leaf using a calibrated stic k 4 EH Ear height cm Measured from the ground suer tfoa cthe node bearing the main ea r 5 EPP Ears per plot Number Number of cobs harv epseter dplot divided by number of plants per plo t 6 SL Stem lodging Number Number of plants with ksst ablroken below the ear per plot 7 RL Root lodging Number Number of plants falleinth w exposed roots per p lot 8 SWT 100 seed weight g Weight of 100 seeds pet r plo 9 SH Shelling % The ratio of grain weight to cob weight expres asse da percentage percentage 10 GT Grain texture Score A scale of 1-5 was used where 1 = very flint (no (1-5) depression at the top of the kernel), 3 = intermated, i5 = very dent (the kernel has a depression) 11 GY Grain yield Kg/ha Total grain yield from athlle ears of each plot with moisture level adjusted to 12.5% and convertedg t/oh ak 12 GLS Gray leaf spot Score Scored using a scale of 1-5, where 1= no disease (1-5) symptoms, 2 = presence of the disease, 3 = mod erate infection, 4 = heavily infested, 5 = severe infoenct i 13 LB Leaf blight Score Scored using a scale of 1-5, where 1= no disease disease (1-5) symptoms, 2 = presence of the disease, 3 = mod erate infection, 4 = heavily infested, 5 = severe infoenct i 14 MSV Maize streak Score Scored using a scale of 1-5, where 1= no disease virus disease (1-5) symptoms, 2 = presence of the disease, 3 = mod erate infection, 4 = heavily infested, 5 = severe infoenct i 15 Rust Rust disease Score Scored using a scale of 1-5, where 1= no disease (1-5) symptoms, 2 = presence of the disease, 3 = mod erate infection, 4 = heavily infested, 5 = severe infoenct i 16 VIG Plant vigour Score Scored using a scale, where 1= healthy, vigoroaunst ,p l (1,3,5) upright; 3 = moderate, 5 = weak, slender or thainn ptsl 17 ASI Anthesis-silking Days The difference between days from planting0 t%o 5pollen interval shedding (AD) and days to 50% silking (DS) i.e. A=S I DS-AD 76 4.4 Data analyses The collected data and derived data were analysiendg uGenStat 1th6 Edition (2013), and Agrobase (2010). Dendrogram construction was cda roriuet in the Number Cruncher Statistical System (NCSS) (Hintze, 2007) using uingwhted pair group method with arithmetic mean (UPGMA). Morphological data were du sand construction was based on Euclidean distance and standard deviation as stycpaele. The genotypic and phenotypic variances were computed from expected mean squoaf rAesN OVA using the formula according to Hallauer and Miranda (1988) as foll:o ws Phenotypic variance denotedσ a2sp σ2 2 2p = σ e + σ g Where :σ2e = error variance an 2 σd g = genotypic variance Genotypic variance is denoted σa2sg σ2g = (MSg - MSe)/r Where: MSg = Mean square of genotypes, MSe = mqeuaanr es error, and r = number of replicates in the experiment. Genotypic coefficient of variation (GCV) GCV = (√σ2g / X) 100 Where,σ 2g= genotypic variance and X = mean of the measured trait Phenotypic coefficient of variation (PCV) PCV = (√σ 2 p / X) 100 Where:σ 2p = Phenotypic variance, X = mean of the trait. Broad sense heritability was computed using thleo wfoinl g equation: H2B = σ2g/ σ2p Where 2 H B is the broa dsense heritabilityσ, 2g = genotypic variance andσ 2p = phenotypic variance. 77 Genetic advance is denoted as GA. GA= k σp H2 Where: k = the standardised selection differenatti a5l % selection (2.063 σ),p = phenotypic standard deviation of the character aHn2 d heritability estimate The across sites correlation coefficients (r) betnw ethe means for grain yield and other agronomic traits were computed in Agrobase (20T1h0e). PCA was carried out using both GenStat Version 16 for latent loadings (Eigen vresc)t oand Agrobase (2010) for Eigenvalues and percent cumulative variation toe gainv insight of the explained variation in the genotypes and traits measured. 4.5 Results All results were reported at a significance levfe Pl ≤o0.05 unless otherwise indicated. 4.5.1 Soil analytical results The pH of the soils and soil nutrients for the ss iatere presented in Table 4.2. Lunyangwa had highly acidic soils (pH 4.50 sub-soil and pH69 4 t.op soils). Tsangano and Bembeke had acidic soils at pH 5.35 and 5.38, 5.13 and f5o.r0 1top soil and sub-soil, respectively. Bvumbwe was classified as having moderately acsiodilcs at pH 5.67 and 5.66 top soil and sub-soil, respectively. The classification was bda soen Soil Test Interpretation Guide (Hornecke t al., 2011) as well as according to Mehlich (1984). At classification of < 0.88 low, 0.88-2.35 mediumnd a 2.35 high (Mehlich III - Mehlich, 1984) all the test sites had medium to low orgacnaircb on. With respect to organic matter (% OM), at classification of 1.5 low, 1.5-4.0 memdi uand > 4.0 high all the sites had medium to low OM. In terms of percent N at clascsaitfion of 0.08 very low 0.08-0.12 low, 0.12-0.2 medium, 0.20-0.30, it was only Bembeke swiht ich was medium in sub-soil (0.16). The rest of the sites were low N. Tsanghaando low P in top soil, medium in sub- soil, Bvumbwe had very high P and Lunyangwa had iumme din top soil and low in sub- soil. Bembeke had low P in both sections of thel sporoi file at classification of P 78 (8 µg g-1 very low, 9-18 µg -g1 low, 19-25 µg -g1 medium (adequate range) > 34 µ-1g v gery high) (Mehlich III - Mehlich, 1984). Table 4.2 Soil characterization for the low pHs s ite Low P Ca Mg pH site Depth (cm) pH %OC %OM %N (µg g-1) (µg g-1) (µg g-1) Al% Tsangano 0-15 5.35 1.23 2.13 0.11 16.75 2.02 0.37 0.40 15-30 5.38 1.97 3.40 0.10 19.27 2.18 0.37 1.33 Bvumbwe 0-15 5.67 1.37 2.36 0.12 64.58 3.38 0.95 0.60 15-30 5.66 1.16 1.99 0.10 62.51 2.81 0.74 0.33 Lunyangwa 0-15 4.69 1.50 2.59 0.13 24.30 1.15 0.21 0.87 15-30 4.51 1.05 1.81 0.09 10.46 0.58 0.11 0.60 Bembeke 0-15 5.13 1.82 3.14 0.16 16.30 2.05 0.42 0.60 15-30 5.01 1.33 2.28 0.11 13.64 1.32 0.25 0.80 OC = organic carbon; OM = organic matter; N = nigtreon; P = phosphorus; Ca = calcium; Mg = magnes ium; Al = aluminium 4.5.2 Combined ANOVA for grain yield and agronotmraicit s at four low pH environments across two seasons 2011/12 and 2012/3 Mean squares from the combined ANOVA across fotuers saind two seasons are given for the top 10 and bottom 10 maize genotypes in terfm gsr aoin yield and agronomic traits (Table 4.3). Results for all entries under low poHil senvironments are presented in Appendix 7. Genotype mean squares were signififcoar ndta ys to anthesis, anthesis-silking interval, days to 50% silking, plant and ear he,i gnhutmber of ears per plant, LB disease, husk cover, GLS disease, grain texture, 100 seeigdh wt aend plant vigour. Location mean squares were significant for all the measured aenridv edd traits. Season mean squares were significant for all characteristics except for ehaerig ht, husk cover, and stem lodging. The GEI (Table 4.3 - GxE) was significant for anthedsaiste , days to 50% silking, ear height, husk cover, grain texture, rotten ears, 100 seeidg hwt,e and plant vigour. Interaction of genotype by season (GxY) mean squares were siagnti ffiocr days to anthesis, days to 50% silking, plant height, number of ears per plant, dLisBease, GLS disease, MSV disease, rust disease, grain texture, rotten ears and 100 seiegdh tw. e 79 The interaction of environment by season (ExY) wnoats s ignificant for husk cover, plant height and rotten ears but was significant for rtehset of the traits. Interaction of genotype by environment by season (GxExY) was significanrt afonthesis date, anthesis-silking interval, days to 50% silking, LB disease and rloodtg ing. 4.5.3 Estimated contributions to total sum of seqsu acrross four low pH soil environments for the 2011/12 and 2012/13 seasons Contribution to total sum of squares for genotypaes wthe highest for husk cover (14.5%) and grain texture (8.4%, Table 4.4). Contributioune dto environment was high for plant vigour (47.1%), grain yield (63.4%), ear height .(74%1) and days to 50% silking (40.6%). The contribution of season was high for rust dies e(a1s8.1%). GxE contribution was significant for grain texture (23.5%) and ear r(o2t7s. 4%). The interaction of genotype by season (GxY) sum of squares made the highest bcuotniotrni to gray leaf spot and MSV (8.8%) and husk cover variation (7.9%). The intteioranc of ExY made the highest contribution to 100 seed weight (22.8%), root londgg (i19.4%) and ear height (14.1%). The interaction of GxExY made the highest contributtion s helling percentage (11.8%) and anthesis date (11.2%) (Table 4.4). 4.5.4. Estimated percent reduction for grain yaienld other salient phenotypic traits at four low pH soil environments versus four opl teimnvaironments across 2011/12 and 2012/13 seasons The reduction in traits was estimated as the deinffceer between the trait’s trial mean for optimal and low pH and subtracted from 100% experde sass a percentage of the optimal i.e. Reduction % = (Xoptimal – Xlowph)/ Xoptimal * 100% . The results are presented in Table 4.5. The combined mean reduction was 69.9%. Thhee hsitg reduction was recorded for shelling percentage (87.5%) followed by number aorfs eper plant (78.0%) and then by plant height (76.6%). 80 Table 4.3 Mean squares for combined ANOVA for g ryaiienld and agronomic traits at four low pH enviroentms across 2011/12 and 2012/13 seasons Source Genotypes Environment Year GxE GxY ExY GYx Ex MSE Df 44 3 1 132 44 3 132 714 GY 4.04E+05 310600000** 65600000** 4.11E+05 5.010E5 + 6821000** 4.59E+05 3.99E+05 AD 60.55** 5708.84** 9122.61** 35.8** 49.45** 15014.** 50.84** 18.05 ASI 5.62** 922.76** 28.34** 3.83 3.67 207.43** 4.5**2 3.2 DS 62.99*** 9903.85** 8172.76** 33.67** 53.27** 2254.89** 45.59** 18.3 EH 292.1* 59649.2** 164.8 310.1** 219.5 30332.6** 3 1.2 185.6 EPP 0.33* 8.71** 35.07** 0.19 0.36* 7.81** 0.24 02. 2 LB 0.88* 36.22** 48.87** 0.71 1.2** 42.24** 0.81* .059 HC 4.99** 36.2** 0.08 3.58** 2.73 1.64 2.72 2.02 GLS 2.47* 20.29** 20.14** 1.94 3.09** 38.11** 1.67 1.59 GT 1.75** 64.85** 47.70** 1.65** 0.85* 3.25** 0.48 0.49 MSV 0.17 17.84** 20.01** 0.14 0.41** 6.63** 0.28 20 . PH 16886** 230489** 274015** 7926 15505* 17503 117 0 10306 RE 1.36 13.14** 45.22** 2.61** 2.12* 1.53 2.63 1. 43 RL 3.28 268.03** 325.41** 3.02 3.77 342.43** 3.65* 2.73 Rust 0.4 53.38** 80.96** 0.36 0.59** 25.82** 0.36 .302 SL 3.06 48.77** 1.63 2.73 2.55 135.27** 2.89 2.5 SWT 60.58** 2419.84** 1117.69** 29.63** 42.7** 339.71** 31.84 20.97 SH 139.2 12739** 1131* 188.4 148.2 908** 209.6 167 6. VIG 1.81** 184.46** 9.42** 0.92** 0.75 49.95** 0.6 6 0.56 ***P ≤0.001; **P ≤0.01; *P≤0.05; G = genotype, E = environment, Y = year, M=S mE ean square error, Df = degree of freedom, GYr a=in g yield (kg h-a1), AD = days to anthesis (days), ASI = anthesis-silking interval (days), DS = daoy si ltking (days), EH= ear height (cm), EPP = eaerrs p plant (#), LB = leaf blight disease (1-5), HChu =s k cover, GLS = gray leaf spot disease (1-5), GT = grain texture (1-5), MSV = mea siztreak virus disease (1-5), PH = plant height) ,( RcmE = rotten ears (#), RL = root lodging (#), Rt =u srust disease (1-5), SL= stem lodging (#), SWT = 100 seed weight (g), SH = snhge lpliercentage, VIG = vigour (1-5), # = number 81 Table 4.4 Estimated percentage contributions taol stoutm of squares for traits at four low pH envmireonts combined for 2011/12 and 2012/13 seasons Variation GY AD ASI DS EH EPP LB GLS GT HC MSV PH RE RL Rust SL SWT SH VIG Genotype 1.2 4.5 3.4 3.8 3.0 5.1 5.2 7.0 8.4 14.5.8 3 6.8 4.8 2.7 3.9 4.7 6.0 2.6 6.8 Environment 63.4 28.6 38.3 40.6 41.7 9.1 9.6 2.6 .4 214.8 17.6 6.4 3.1 15.2 23.8 5.1 16.2 16.4 47.1 Year 4.5 15.3 0.4 11.2 0.0 12.2 6.5 1.3 5.2 0.0 9.9 2.5 3.6 6.1 18.1 0.1 2.5 0.5 0.8 GxE 3.7 7.9 7.0 6.1 9.5 8.7 8.3 11.1 23.5 20.8 6.2 9.6 27.4 7.5 7.1 12.6 8.7 10.6 10.4 GxY 1.5 3.6 2.2 3.2 2.3 5.5 7.2 8.8 4.1 7.9 8.8 6.37.4 3.1 5.9 3.9 4.2 2.8 2.8 ExY 1.4 7.5 8.6 9.2 14.1 5.4 11.2 4.9 0.7 0.0 3.3 .3 0 1.2 19.4 5.8 9.5 22.7 1.2 4.3 GxExY 4.1 11.2 8.3 8.2 2.7 7.2 9.4 9.5 4.7 0.3 6.0 9.5 1.6 9.1 3.5 8.9 9.1 11.8 2.5 Residual 19.4 21.4 31.6 17.7 26.6 46.7 41.8 54.8 .7 3147.6 44.0 58.3 50.9 36.8 31.6 54.7 29.7 53.7 5 25. Rep 0.8 0.0 0.2 0.0 0.0 0.3 0.8 0.6 0.2 4.2 0.5 0.30.0 0.1 0.3 0.4 1.0 0.4 0.1 CV (%) 42.4 5.1 67.1 5.0 28.1 54.2 35.2 82.9 58.9 7.82 36.5 77.0 102.5 92.1 44.6 127.6 20.3 19.2 27.4 Min 0.8 0.0 0.2 0.0 0.0 0.3 0.8 0.6 0.2 0.0 0.5 0.30.0 0.1 0.3 0.1 1.0 0.4 0.1 Max 63.4 28.6 38.3 40.6 41.7 46.7 41.8 54.7 31.7 .6 4744.0 58.3 50.9 36.8 31.6 54.7 29.7 53.7 47.1 G = Genotype, E = environment, Y = year, CVc o= efficient of variatio, nMin = minimum, Max = maximum, GY = grain yieldg ( kha-1), AD = days to anthesis (days), ASI = anthesis- silking interval (days), DS = days to silking ()d, aEyHs= ear height (cm), EPP = ears per plant (#B), =L leaf blight disease (1-5), GLS = gray leaf sdpisoetase (1-5), GT = grain texture (1-5), HC = husk cover, MSV = maize streak viruse daise (1-5), PH = plant height (c mRE), = rotten ears (#), RL = root lodging (#), Rus rtu =st disease (1-5), SL= stem lodging (#), SWT = 100 seed weight (g), SH = shelling percen, tVaIgGe = vigour (1-5), # = number 82 Table 4.5 Estimated reduction of grain yield anhde ro st alient traits under low pH versus optimal conditions across the 2011/12 a1n2d/1 230 seasons Trait Optimal Low pH % Reduction GY 3372.9 1490.3 44.2 EH 77.6 46.8 60.3 EPP 1.1 0.9 78.0 PH 172.3 131.9 76.6 SWT 29.9 22.6 75.5 SH 79.2 69.3 87.5 VIG 4.0 2.7 67.5 Mean 69.9 GY = grain yield (kg h-a1), EH = ear height (cm), EPP = ears per plant (#), =P Hplant height (cm), SWT = 100 seed weight (g), SH = shelling percentage, VIGg o= uvri (1-5), # = number 4.5.5 Genotypic and phenotypic variance compon geenntse,tic advance and broad sense heritability estimates across four low plH e nsovironments combined for 2011/12 and 2012/13 seasons The results in Table 4.6 indicated that phenotyvpaircia nces were higher than genotypic variances. Grain texture (0.56), days to 50% sgil k(i0n.55), anthesis date (0.54) and plant vigour (0.53) had relatively high broad sense haebriliitty estimates. The genetic coefficients of variations were lower than phenotypic coeffictsie onf variation except for plant height. The expected genetic advance was highest for hpelaignht t followed by ear height and these were followed by anthesis date and 100 seed we Gigrhati.n yield had the lowest expected genetic advance with high phenotypic coefficien vt aorfiation (Table 4.7). 83 Table 4.6 Genotypic variances, phenotypic varia anncde sheritability estimates at low pH sites across two seasons 2011/12 and 2 012/13 Trait σ2g σ2 H2p b H2b % GY 2.300 401.400 0.006 0.623 AD 21.250 39.300 0.541 54.071 ASI 1.200 4.400 0.273 27.273 DS 22.345 40.645 0.550 54.976 EH 53.250 238.850 0.223 22.294 EPP 0.055 0.275 0.200 20.000 LB 0.145 0.735 0.197 19.728 HC 0.450 2.050 0.220 21.951 GLS 0.630 1.120 0.563 56.250 GT 1.485 3.505 0.424 42.368 MSV 0.015 0.185 0.081 8.108 PH 3290.000 13596.000 0.242 24.198 RE 0.035 1.395 0.025 2.509 RL 0.540 2.740 0.197 19.708 Rust 0.040 0.360 0.111 11.111 SL 0.280 2.780 0.101 10.072 SWT 19.815 40.785 0.486 48.584 SH 18.700 157.900 0.118 11.843 VIG 0.620 1.180 0.525 52.542 Min 0.015 0.185 0.081 0.623 Max 3290.000 13596.000 0.563 56.250 Min = minimum, Max = maximum, GY = grain yield (hkga- 1), AD = days to anthesis (days), ASI = antheskisin-sgi l interval (days), DS = days to silking (days), EHr= h eaight (cm), EPP = ears per plant (#), LB = leaigf hbtl disease (1-5), HC = husk cover, GLS = gray leaf spot dis e(1a-s5e), GT = grain texture (1-5), MSV = maize strke vairus disease (1-5), PH = plant height (c mRE), = rotten ears (#), RL = root lodging (#), Rustu =s tr disease (1-5), SL= stem lodging (#), SWT = 100 seed weight (g), SHe =lli nshg percentage, VIG = vigour (1-5), # = number 84 Table 4.7 Genotypic coefficient of variation, phtyepnico coefficient of variation and expected genetic advance at low pH sites acro sse atwsons 2011/12 and 2012/13 Trait GCV% PCV% GA GA (% of mean) GY 12.4 164.1 0.5 32.7 AD 5.1 6.9 2.6 3.2 ASI 6.7 12.9 1.1 41.0 DS 5.1 6.9 2.7 3.1 EH 10.7 22.6 2.7 5.7 EPP 2.6 5.6 0.5 54.8 LB 2.6 5.8 0.6 26.9 HC 9.3 14.3 1.3 74.6 GLS 4.5 9.7 0.8 37.4 GT 5.0 6.7 1.1 43.7 MSV 1.0 4.2 0.2 19.8 PH 49.9 101.5 7.6 5.8 RE 1.8 11.2 0.2 21.3 RL 3.9 12.9 0.6 31.9 Rust 1.8 5.3 0.4 29.4 SL 4.8 15.0 0.6 47.7 SWT 9.4 13.4 2.5 11.2 SH 5.2 16.8 1.7 2.4 VIG 4.8 6.6 1.1 39.7 Min 1.0 4.2 0.2 2.4 Max 49.9 164.1 7.6 74.6 GCV = genotypic coefficient of variation, PCV = npohteypic coefficient of variation, GA = genetic nacdev,a GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (days), EH= ear height (cm), EPP = ears per plan,t L(#B) = leaf blight disease (1-5), HC = husk coveLrS, G= gray leaf spot disease (1-5), GT = grain texture (1-5S),V M = maize streak virus disease (1-5), PH = plaenigt ht (cm) , RE = rotten ears (#), RL = root lodging (#), Rustu =s tr disease (1-5), SL= stem lodging (#), SWT = 1e0d0 wseight (g), SH = shelling percentage, VIG = vigour (1#-5 =), number, Min = minimum, Max = maximum. 4.5.6 Mean performance for grain yield and othearit str across four low pH soil environments combined for 2011/12 and 2012/13 nsse aso The grain yield trial mean was 1490 kg- 1h (aTable 4.8) and the top five performing genotypes were G24 (LPHpop16) G30 (LPHpop3) G22O (7V3P9), G15 (VPO5173) and G27 (LOW N POOL C3-B). The latest maturing genot ywpaes G35 (86 AD = 172 days 85 Table 4.8 Mean performance for grain yield and or tahgeronomic traits across four low pH environmceonmtsb ined for 2011/12 and 2012/13 seasons Entry GY AD ASI DS EH EPP LB HC GLS GT MSV PH RE RL Rust SL SWT SH% VIG Top 10 genotypes G24 1792 79.7 3.0 83.0 51.2 0.8 2.4 1.6 .3 2 2.4 1.1 134 1.4 2.3 1.5 1.2 24 73 2.7 G30 1765 82.5 2.9 85.4 44.0 0.9 1.9 2.4 2.1 2.3 3 1. 119 1.1 2.0 1.3 0.6 24 74 2.1 G22 1753 81.6 2.5 84.8 48.2 0.9 2.2 2.1 2.1 2.4 2 1. 126 1.2 1.7 1.2 1.1 23 70 2.7 G15 1746 82.5 3.5 86.0 48.6 0.9 2.3 1.7 2.4 2.7 1 1. 132 1.2 2.2 1.5 1.4 22 70 2.8 G27 1701 84.7 2.0 86.7 49.8 0.9 2.1 1.0 1.9 2.8 2 1. 129 1.2 1.3 1.1 1.7 27 69 2.4 G6 1655 82.0 2.0 83.1 44.4 0.9 2.4 1.3 2.7 2.8 1.2129 1.6 1.4 1.2 0.7 22 68 2.7 G16 1650 80.8 3.1 82.9 38.9 0.8 1.9 1.9 2.1 2.6 3 1. 116 1.0 1.3 1.4 1.6 24 71 2.7 G21 1602 83.8 2.8 86.6 42.6 0.9 2.3 1.7 2.0 2.9 1 1. 126 1.0 1.8 1.4 1.0 22 71 2.6 G34 1599 81.6 3.0 84.0 44.7 0.8 2.0 1.1 1.8 1.8 3 1. 123 0.9 1.7 1.3 1.1 22 70 2.7 G26 1575 84.5 3.2 87.7 47.5 0.8 2.6 2.2 2.1 1.9 3 1. 125 1.2 1.3 1.2 0.9 23 68 2.7 Bottom 10 genotypes G25 1401 84.5 2.7 86.4 50.8 0.9 2.2 1.5 .9 1 2.5 1.2 141 1.1 1.5 1.3 1.1 23 69 2.5 G35 1396 86.3 3.0 89.3 45.2 0.8 2.1 2.3 2.3 3.2 1 1. 126 1.4 2.0 1.2 1.1 26 69 2.8 G33 1385 83.0 3.0 86.5 46.6 0.8 2.0 1.2 2.1 2.5 2 1. 127 1.0 2.4 1.3 1.3 22 70 2.5 G41 1376 80.3 2.8 83.5 40.6 0.8 2.0 1.9 3.7 2.6 3 1. 118 1.0 1.8 1.2 1.0 21 67 2.8 G4 1372 83.5 3.4 87.3 48.6 0.8 2.3 2.2 2.2 2.7 1.1127 1.7 1.8 1.4 0.9 23 68 2.4 G14 1362 83.6 2.7 86.3 46.5 0.8 2.2 2.5 2.3 2.3 2 1. 148 0.9 2.0 1.0 1.3 22 68 2.7 G44 1359 83.9 2.4 86.7 48.1 0.9 2.1 1.2 2.1 2.4 3 1. 135 0.9 1.8 1.4 1.0 23 67 3.3 G10 1314 84.6 2.8 87.5 46.2 1.0 2.3 1.4 2.1 2.9 3 1. 123 0.9 2.3 1.4 1.3 19 68 2.5 G2 1282 82.4 2.3 84.8 48.2 0.8 2.4 1.6 1.8 2.4 1.2127 0.9 2.5 1.4 1.7 22 65 3.0 G31 1195 80.3 2.6 82.9 45.4 0.9 2.0 2.3 2.1 2.6 3 1. 123 1.3 2.2 1.3 0.9 22 64 3.1 Mean 1490.6 83.1 2.7 85.8 46.8 0.9 2.2 1.7 2.1 2.51.2 131.9 1.2 1.8 1.3 1.2 22.6 69.3 2.7 LSD 358 2.4 1 2.4 7.7 0.3 0.4 0.8 0.7 0.4 0.3 58 .7 0 0.9 0.3 0.9 3 8 0.4 CV (%) 42.4 5.1 67 5 28 54 35 83 59 28 37 77 103 2.19 44.6 127.6 20.3 19 27.4 SE 631 4.2 1.8 4.3 13.6 0.5 0.8 1.4 1.3 0.7 0.5 1.51 0 1.2 1.7 0.6 1.6 4.6 13 0.8 Min 1195.0 79.7 1.6 82.9 38.0 0.7 1.8 0.8 1.6 1.8 1.1 116.0 0.8 1.2 1.0 0.6 19.0 64.0 2.1 Max 1792.0 86.3 3.5 89.3 53.9 1.0 2.7 2.6 3.7 3.2 1.5 148 1.7 2.7 1.5 2.2 27.0 75.0 3.3 LSD = Least significant difference, CV = coeffitc oief nvariation, SE = Min = minimum, Max = maximuGmY, = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (days), EH= ear height (cm), E=P ePa rs per plant (#), LB = leaf blight disease) ,( 1H-5C = husk cover, GLS = gray leaf spot disea-s5e), (G1T = grain texture (1-5), MSV = maize streaikr uvs disease (1-5), PH = plant height (c mRE), = rotten ears (#), RL = root lodging (#), Ru srtu =st disease (1-5), SL= stem lodging (#), SWT0 = s 1e0ed weight (g), SH = shelling percentage, VvIGig o=u r (1-5), # = number 86 physiological maturity period) and the earliestt hoef group was G24 (80 AD = 160 days physiological maturity period). Maturity period faonr y maize genotype is twice the number of days to flowering. The genotype most suscep ttiob legray leaf spot disease was G 41. The shortest genotype as a result of low pH efwfeacst G 16 (116 cm). Genotype G27 recorded the highest 100 seed weight (27 g) foldlo wbye G35 (26 g). The best shelling percentage was recorded for G30 (74%). The gen owtyitphe the best plant vigour was G30 followed by G27 and G4 (Table 4.8). 4.5.7 Pearson’s correlation coefficients betweaein ygireld and agronomic traits across four low pH soil environments combined0 f1o1r/ 122 and 2012/13 seasons Grain yield was positively and highly significan tlcyorrelated with anthesis date (r = 0.12), anthesis-silking interval (r = 0.35a),y ds to silking (r = 0.21), number of ears per plant (r = 0.17) and shelling percentage (r = 0.. 1D9a)ys to anthesis were significantly correlated with most traits excluding stem lodgainngd it was negatively correlated with plant and ear height as well as root lodging. Asnitsh-esilking interval also correlated with traits like: days to silking, shelling percentagned aears per plant. Days to silking correlated negatively with plant and ear height (-0.20 and4 0-0) .as well as with root lodging (-0.10). Correlation for plant height was positively withr ehaeight, husk cover and 100 seed weight, 0.40, 0.10 and 0.2, respectively. Ear height caotreredl with traits like, root lodging, 100 seed weight, ears per plant, shelling percentaigoer, vand diseases such as leaf blight and rust. Number of ears per plant furthermore coreresl awtith among others, grain texture, 100 seed weight and vigor (Table 4.9). 4.5.8 Principal component analysis results, eiglueensv and eigenvectors for the traits across four low pH soil environments comd bfoinre 2011/12 and 2012/13 seasons Results indicated that eight principal componenetsr ew generated (Table 4.10) which accounted for 100% variability present in the m agiezneotypes evaluated. The first five PC had eigenvalues higher than 1 and their cumulapteivrcee ntages accounted for 71.1% of the total variation present among the genotypes. 87 Table 4.9 Pearson's correlation coefficients faoirn g yrield and agronomic traits across four low npvHir oenments for two seasons GY AD ASI DS PH EH RL SL HC RE GT LB GLS Rust MSV SWT VIG SH AD 0.12** ASI 0.35** 0.30** DS 0.12** 0.90** 0.30** PH -0.02 -0.10** -0.02 -0.20** EH -0.03 -0.30** -0.01 -0.40** 0.40** RL 0.02 -0.10** 0.03 -0.10** 0.07 0.34** SL 0.03 0.04 0.01 0.03 -0.03 -0.01 0.04 HC -0.02 0.10** -0.01 0.04 0.10** 0.08 -0.20** 0.02 RE -0.06 0.20** -0.03 0.20** -0.03 -0.20** -0.20** 0.01** 0.30** GT 0.03 0.09** 0.02 0.12** 0.01 -0.02 0.00 -0.04 -0*.*1 0 0.04 LB 0.01 0.12** 0.02 0.09** 0.06 0.09** 0.12** 0.17** .08 0.22** 0.30** GLS -0.02 0.14** 0.00 0.14** 0.04 0.03 0.13** 0.03 0. 00 0.16** 0.30** 0.50** Rust -0.08 0.08* -0.01 0.01 0.16 0.11** -0.08 0.15** 0.*3* 0.34** 0.30** 0.60** 0.40** MSV -0.02 0.08* -0.01 0.02 0.06 0.04 -0.03 0.20** 0.*1 3* 0.29** 0.30** 0.70** 0.40** 0.60** SWT 0.05 0.06* 0.04 0.05 0.20** 0.40** 0.38** 0.05 -03. 0 0.01 -0.03 0.20** 0.20** 0.04 0.00 VIG 0.03 0.13** 0.02 0.16** 0.01 0.09** 0.29** 0.12** 0-.07 0.14** 0.10** 0.30** 0.20** 0.20** 0.10 0.40* * SH 0.19** 0.27** 0.30** 0.24** 0.05 0.09** 0.07 0.05 .05 0.06 0.10** 0.10** 0.10** 0.10 0.10 0.20** 0.1**0 EPP 0.17** 0.67** 0.30** 0.65** 0.07 0.12** 0.07 0.05 .05 0.06 0.10** 0.10** 0.10** 0.10** 0.10** 0.20** 0.10** 0.40* **P ≤0.01, * P≤0.05, GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daPyHs) =, plant height (cm), EH= ear height (cm), RL = root lodging (#), SL= stem lodging H(#C) ,= husk cover, RE = rotten ears (#), GT = graeinx tture (1-5), LB = leaf blight disease (1-5), GL gSr a=y leaf spot disease (1-5), Rust = rust disease (1-5), MSV = maize streak v diriussease (1-5), SWT = 100 seed weight (g), VIGg =o uvri (1-5), SH = shelling percentage, EPP = ear sp lpaent (#), (#) = number. 88 Table 4.10 Eigenvalues, percentages and cumulpaetirvce ntages for the measured and derived data across four low pH soil environtsm ceonmbined for 2011/12 and 2012/13 seasons Principal component Eigenvalues As percentages lCatuivmeu percentages 1 2.2 18.5 18.5 2 1.8 15.7 34.2 3 1.5 13.1 47.4 4 1.5 12.5 59.8 5 1.3 11.3 71.1 6 1.2 10.0 81.1 7 1.1 9.7 90.8 8 1.1 9.2 100.0 4.5.9 Clustering of maize genotypes evaluatedu ar tl ofow pH soil environments combined across 2011/12 and 2012/13 seasons The dendrogram was constructed using the UPGMAt ecrl uasnalysis method. Forty five maize genotypes were clustered based on meanrsa fionr ygield. At cut off point 1.0, three main clusters were observed (Figure 4.1) with ah hciogphenetic correlation ocfo rp = 0.82. Thirteen genotypes from CIMMYT-Zimbabwe were grodu pine cluster two with only two from CIMMYT-Colombia. The top performing genotypwese re grouped in cluster three which comprised of 24 genotypes from both Colomabniad Zimbabwe. Cluster one was comprised of six genotypes which were also fromh bCoItMMYT research centres. 89 G24 G30 G22 G15 G27 G16 G6 G31 G10 G2 G40 G20 G23 G19 G11 G29 G13 G8 G18 G9 G7 G28 G5 G34 G21 G26 G3 G4 G45 G32 G39 G1 38 G36 G17 G42 G37 G35 G25 G43 G12 G1 2.50 2.08 1.67 1.25 0.83 0.42 0.00 Dissimilarity Figure 4.1 Dendrogram based on Euclidean distandc eU aPGMA clustering using morphological data fonro gtyepes at four low pH environments combined for 2011/12 and 2012/13 nsse aso 90 4.5.10 Performance of maize genotypes across pfotiumr aol soil environments combined for 2011/12 and 2012/13 seasons Mean squares from the combined ANOVA across fotuers saind across two years presented are for the top10 and bottom 10 maize genotypetse rimn s of grain yield (Table 4.11). Results for all entries under optimal environmeanrets p resented in Appendix 7. Significant differences due to genotypes were observed for e1a6s mured traits except for anthesis- silking interval, husk cover and root lodging. Ernovniment mean squares were significant for all traits except for 100 seed weight. Seasoena nm squares were significant for all traits excluding MSV and 100 seed weight. G x E mean seqsu warere significant for most traits except for grain yield, anthesis-silking intervnaul,m ber of ears per plant, leaf blight and gray leaf spot disease as well as shelling pergcen. tGa x Y interaction mean squares were not significant for anthesis-silking interval, mea isztreak virus, rotten ears and rust disease. The interaction of E x Y mean squares was signnifti cfoar traits except for diseases: leaf blight, MSV and rust, and the following traits: hku csover, stem and root lodging, rotten ears, plant vigour and 100 seed weight. The intieorna cG x E x Y was only significant for grain yield, anthesis date, days to 50% silkinga,y g ler af spot, grain texture and plant vigour (Table 4.11). 4.5.11. Estimated contributions to total sum oaf rseqsu across four optimal environments combined for 2011/12 and 2012/13 nsse aso Contribution due to genotype was the highest fo0r s1e0ed weight (14.1%), MSV (13.4%) and stem lodging (12.0%) (Table 4.12). Contribu tdioune to environment was high for anthesis date (67.0%), plant vigour (53.7%), Ru4s8t. 1(%), ear height (48.2%). The contribution due to the effect of season was hiogrh r of ot lodging (24.1%), rust disease (18.5%) and rotten ears (16.5%). G x E contribu ttio nvariation was high for MSV (19.1%), husk cover (21.1%) and stem lodging (12). 5T%he interaction of genotype by season sum of squares made the highest contrib tuot isotnem lodging (22.8%), root lodging (14.2%) and 100 seed weight (65.7%). The interna ctoiof E x Y made the highest contribution to gray leaf spot (12.5%) anthesise d(a1t4.1%) and days to 50% silking 91 Table 4.11 Mean squares for combined ANOVA for ng ryaiei ld and agronomic traits at four optimal ennvmireonts for 2011/12 and 2012/13 seasons Source Genotypes Environment Year GxE GxY ExY GYx Ex MSE Df 44 7 1 307 44 6 264 1334 GY 5753000*** 1.1E+08** 4E+07** 1.33E+06 4978000 ** 21230000** 1597000** 1.15E+06 AD 117.10** 22428.24** 3497.57** 22.12** 48.65** 7024.79** 18.78** 8.70 ASI 3.30 442.03** 437.178** 5.00 6.20 207.96** 5. 30 4.90 DS 111.87** 20256.23** 6094.80** 20.60** 49.30** 6801.36** 20.66** 9.30 EH 690.00** 77831** 3466.30** 270.1** 457.30** 802.380** 236.10 199.10 EPP 0.07** 0.40** 1.22** 0.03 0.0812** 0.65** 0.03 0.03 LB 0.80** 79.38** 50.09** 0.32 1.01** 6.44 43.10 207. HC 2.70 326.10** 9.44* 4.55** 4.10** 4.40 23.10 20. 1 GLS 0.93** 41.80** 1.85* 0.30 0.96** 63.96** 0.55* 0.34 GT 2.40** 55.50** 45.90** 1.10* 1.60** 8.67** 1.03 * 0.80 MSV 0.60** 35.28** 0.00 0.40** 0.00 5.41 32.10 0. 20 PH 1277.50** 101791.30** 29780.1** 557.8* 933.30** 21985.9** 490.80 432.80 RE 1.18** 85.10** 80.96** 0.67** 0.30 5.20 22.10 400. RL 4.20 1571.03** 2323.97** 10.51** 31.02** 6.41 4.120 4.30 Rust 0.80** 118.80** 137.14** 0.45** 0.20 3.40 210. 1 0.30 SL 3.94** 128.45** 8.80** 2.05** 7.48** 2.40 29.20 1.30 SWT 80.30** 145.60 50.80 29.20** 89.86** 2157.90 .30 25.40 SH 241.80** 3012.40** 2294.20** 119.90 174.20** 185.310** 100.60 108.00 VIG 0.70** 138.75** 6.65** 0.53** 1.45** 0.60 0.53* 0.30 **P ≤0.01; *P≤0.05; G = genotype, E = environment, Y = year, M=S Em ean square error, Df = degree of freedom, GYr a=i ng yield (kg h-a1), AD = days to anthesis (days), ASI = anthesis-silking interval (days), DS = days toin sgi l(kdays), EH= ear height (cm), EPP = ears pern pt l(a#), LB = leaf blight disease (1-5), HC = huosvke cr, GLS = gray leaf spot disease (1-5), GT = grain texture (1-5), MSV = maize stre vaikrus disease (1-5), PH = plant height (c RmE) ,= rotten ears (#), RL = root lodging (#), Ru srtu =st disease (1-5), SL= stem lodging (#), SWT = 100 seed weight (g), SH = shelling pnetracgee, VIG = vigour (1-5), # = number 92 Table 4.12 Estimated percent contributions to tsoutmal of squares at four optimal environments a c2r0o1s1s/12 and 2012/13 seasons Source GY AD ASI DS EH EPP LB GLS GT HC MSV PH RE LR Rust SL SWT SH VIG Genotypes 11.9 5.0 2.2 5.1 6.3 7.5 6.6 7.8 9.3 4.31 3.4 6.5 10.6 1.9 4.8 12.0 14.1 7.6 4.0 Environments 15.6 65.7 19.5 63.0 48.2 3.1 44.6 24.124.7 34.6 38.5 35.2 34.7 32.6 48.1 17.8 1.2 6.4 .7 53 Year 1.8 3.4 6.4 6.3 0.7 3.1 9.4 0.4 4.0 0.3 0.0 4 3. 16.5 24.1 18.5 0.6 0.2 1.6 0.9 GxE 8.3 2.8 9.7 2.8 7.3 10.9 7.9 9.8 12.4 21.1 19.81.4 12.1 9.6 7.9 12.5 10.2 11.3 9.0 GxY 10.3 2.1 4.0 2.3 4.2 9.2 8.3 8.1 6.3 6.4 0.9 7 4. 2.8 14.2 1.5 22.8 15.7 5.5 8.2 ExY 3.0 14.1 6.1 12.6 3.3 5.0 0.0 12.3 1.5 0.0 0.0 5.1 0.1 1.0 1.8 0.1 8.6 3.3 0.1 GxExY 9.9 1.6 6.8 1.9 4.3 9.3 0.7 4.7 8.0 1.0 0.0 .0 5 0.0 0.3 0.2 0.5 5.3 9.4 3.0 Rep 0.6 0.0 0.2 0.0 0.0 0.7 0.3 0.5 0.1 0.0 0.1 0.60.3 0.2 1.0 0.5 0.4 0.0 0.1 Residual 38.7 5.3 45.1 6.0 25.7 51.3 22.2 32.2 43.372.4 28.0 31.1 22.9 16.1 16.2 33.1 44.4 54.9 21.1 CV (%) 31.8 4.6 194.2 4.6 18.2 15.2 27.1 103.1 3 6.0103 35.9 12.1 80.3 301.5 31.3 169.4 16.8 13.1 27.3 Min 0.6 0.0 0.2 0.0 0.0 0.7 0.0 0.4 0.1 0.0 0.0 0.60.0 0.2 0.2 0.1 0.2 0.0 0.1 Max 38.7 65.7 45.1 63.0 48.2 51.3 44.6 32.2 43.7 .6 34 38.5 35.2 34.7 32.6 48.1 33.1 44.4 54.9 53.7 G = genotype, E = environment, Y = year, C Vc o=efficient of variatio nM,in = minimum, Max = maximum, GY = grain yield ( khga-1), AD = days to anthesis (days), ASI = anthesis- silking interval (days), DS = days to silking ()d, aEyHs = ear height (cm), EPP = ears per plant (L#B), = leaf blight disease (1-5), GLS = gray leaft sdpisoease (1-5), GT = grain texture (1-5), HC = husk cover, MSV = maize streak viruse daise (1-5), PH = plant height (c mRE), = rotten ears (#), RL = root lodging (#), Ru srtu =st disease (1-5), SL= stem lodging#, SWT = 100 seed weight (g), SH = shelling percentagGe ,= V vIigour (1-5), # = number 93 (12.6%). The interaction of G x E x Y made the heisgth contribution to grain yield (9.9%), shelling percentage (9.4%) and number rosf peear pant (9.3%) 4.5.12 Genotypic and phenotypic variance compo,n geenntestic advance and broad sense heritability estimates across four optimviarlo ennments combined for 2011/12 and 2012/13 seasons The results in Table 4.13 indicated that phenot yvpairciances were higher than genotypic variances. Anthesis date (0.86), days to 50% sgi lk(i0n.85) and grain yield (0.67) had relatively high broad sense heritability estima tReoso. t lodging had the lowest heritability estimate (0.02). The genetic coefficient of vaorina twi as lower than phenotypic coefficient of variation. The expected genetic advance (GA) twhaes highest for grain yield (68.9) followed by plant height (29.5) and ear height 5(2).2 .It was low for number of ears per plant (0.2) and root lodging (0.4) (Table 4.14). Table 4.13 Genotypicσ 2(g) and phenotypicσ (2p) variances and broad sense2 b(H) heritability estimates at four optimal environm eanctrsoss 2011/12 and 2012/13 seasons Trait σ2g σ2p H2b H2b % GY 230.20 345.00 0.67 66.71 AD 54.19 62.90 0.86 86.14 ASI 0.77 4.11 0.19 18.82 DS 51.29 60.58 0.85 84.67 EH 245.45 444.60 0.55 55.21 EPP 0.02 0.05 0.41 40.70 LB 0.27 0.53 0.50 49.98 HC 0.34 2.40 0.14 13.98 GLS 0.29 0.64 0.46 46.20 GT 0.79 1.61 0.50 49.51 MSV 0.20 0.40 0.52 51.63 PH 422.40 855.20 0.49 49.39 RE 0.38 0.80 0.48 47.68 RL 0.08 4.24 0.02 1.97 Rust 0.26 0.54 0.49 49.05 SL 1.30 2.64 0.49 49.44 SWT 27.47 52.80 0.52 52.00 SH 66.90 174.90 0.38 38.25 VIG 0.20 0.50 0.39 39.28 σ2g = genotypic varianceσ,2 p = phenotypic variance, 2bH = broad sense heritability,2 bH % = broad sense heritability as percentage, GY = grain yield (kg-1) h, Aa D = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (days), EH = ear he (icgmht), EPP = ears per plant (#), LB = leaf blighte daisse (1-5), HC = husk cover, GLS = gray leaf spot disease (G1-T5 )=, grain texture (1-5), MSV = maize streak virduis ease (1-5), PH = plant height (cm R),E = rotten ears (#), RL = root lodging (#), Rustu =s tr disease (1-5), SL= stem lodging (#), SWT = 100 seed weight (g), SH = shge pllienrcentage, VIG = vigour (1-5), # = number. 94 Table 4.14 Genotypic coefficient of variation, pohtyepnic coefficient of variation and expected genetic advance at optimal combined 1fo1r/1 220 and 2012/13 seasons Trait GCV% PCV% GA GA % of mean GY 8.3 10.1 68.9 2.4 AD 9.2 9.9 10.6 21.8 ASI 8.2 19.0 1.3 69.2 DS 8.8 9.6 10.3 20.5 EH 17.8 23.9 22.5 31.0 EPP 1.3 2.1 0.2 16.6 LB 3.7 5.3 0.7 39.5 HC 4.9 13.1 0.8 32.1 GLS 4.3 6.3 0.8 46.8 GT 6.5 9.2 1.3 68.4 MSV 4.1 5.7 0.6 55.0 PH 15.7 22.3 29.5 17.3 RE 6.9 10.0 0.9 109.2 RL 3.5 24.8 0.4 12.1 Rust 4.0 5.7 0.7 44.4 SL 13.8 19.7 1.6 243.0 SWT 9.6 13.3 7.5 26.1 SH 9.2 14.9 11.7 13.2 VIG 3.1 5.0 0.6 28.4 GCV = genotypic coefficient of variation, PCV =n pohteypic coefficient of variation, GA = genetic nacdev,a GY = grain yield (kg h-a1), AD = days to anthesis, ASI = anthesis-silkitnegr vinal (days), DS = days to silking (days), EH= ear height (cm), EPP = ears per plant (#), LB = lbelaigf ht disease (1-5), HC = husk cover, GLS = grayf slepot disease (1-5), GT = grain texture (1-5), MSV = ma siztreeak virus disease (1-5), PH = plant height ,( cRmE) = rotten ears (#), RL = root lodging (#), Rust = rduiste ase (1-5), SL= stem lodging (#), SWT = 100 seeigdh wt (g), SH = shelling percentage, VIG = vigour (1-5), #u =m nber. 4.5.13 Mean performance for grain yield and othraeirt st across four optimal environments combined for 2011/12 and 2012/13 nsse aso The grain yield trial mean was 3.3 t- 1h and the top seven performing genotypes were G20 (4.5 t ha-1), G33 (4.0 t h-a1), G8 (4.0 t h-a1), G27 (4.0 t h-a1), G5 (3.9 t h-a1), G31 (3.8 t h-a 1) and G10 (3.8 t h-1a) (Table 4.15). The lowest yielding were G40 (2 h.3a -1t) and G2 (2.3 t ha-1). The earliest maturing genotype was G24 (59 AeD. 1 i.18 days maturity period) and the shortest genotype was G2 (156 cm). The tagllensot type was G5 (189 cm). The best shelling percentage was recorded for G20 (83%) ew hGil2 had the lowest shelling percentage. The genotype with the best plant v igwoausr G20 (1.5). High 100 seed weight was recorded for G40 (33.2 g). 95 Table 4.15 Mean performance for grain yield acrfosusr optimal environments combined for 2011/122 a0n1d2 /13 seaso ns Entry GY AD ASI DS EH EPP LB HC GLS GT MSV PH RE RL Rust SL SWT SH VIG Top ten genotypes G20 4489 65.0 1.0 66.7 81 1.1 1.9 1.3 1.51.8 1.2 173 1.1 1.4 1.3 0.4 32.0 83.0 1.5 G33 4003 61.3 1.2 63.4 75 1.2 1.8 1.5 1.4 1.9 1.6 168 0.7 1.0 1.7 0.9 29.5 82.1 1.9 G8 3965 65.9 1.1 67.6 85 1.2 1.9 1.6 1.9 1.7 1.2 82 1 0.7 1.4 2.0 1.2 30.1 80.9 1.9 G27 3965 66.4 1.4 68.4 82 1.2 1.9 1.6 1.7 2.1 1.2 179 0.8 0.3 1.6 -0.1 30.7 79.8 1.9 G5 3921 68.3 1.0 70.0 87 1.1 1.9 1.9 1.5 1.9 1.1 89 1 0.7 0.8 1.4 1.0 32.1 79.6 1.9 G39 3902 64.1 0.6 65.6 70 1.1 2.0 1.2 1.5 1.9 1.1 161 0.9 0.6 1.8 0.8 31.4 79.3 2.1 G31 3820 64.8 1.0 66.6 82 1.1 2.2 1.3 1.8 1.8 1.2 176 0.8 0.3 1.7 1.0 27.6 80.4 2.0 G10 3817 67.7 2.0 70.2 86 1.1 2.0 0.9 1.9 2.0 1.2 183 0.8 0.6 1.7 0.2 31.1 81.8 2.0 G36 3757 62.9 1.1 64.9 83 1.1 1.9 1.3 1.6 2.0 1.3 179 0.8 0.8 1.9 0.7 28.6 79.5 2.0 G43 3750 61.9 0.9 63.8 69 1.1 2.0 1.3 1.6 1.9 1.2 162 0.9 0.2 1.7 1.7 31.0 77.4 1.9 Bottom ten genotypes G3 3014 64.1 0.8 65.7 75 1.2 1.4 1.3 1.71 .6 1.3 163 0.7 0.5 1.5 1.0 30.1 76.4 2.0 G42 2996 63.2 1.0 64.9 71 1.0 1.8 0.9 1.6 1.7 1.2 168 0.5 1.6 1.7 0.5 31.3 78.7 2.2 G24 2983 59.1 1.3 61.1 72 1.1 2.2 1.8 1.3 2.0 1.1 167 0.7 0.7 1.7 0.8 29.1 75.9 2.2 G16 2901 63.2 0.8 64.9 80 1.0 1.8 2.1 1.8 1.8 1.1 173 0.8 0.5 1.8 0.6 32.3 77.7 2.0 G7 2663 65.6 1.4 67.6 78 1.1 1.9 0.9 1.4 2.0 1.3 71 1 0.8 0.7 1.5 0.6 30.9 80.0 2.0 G37 2594 65.3 1.3 67.2 66 1.1 1.7 1.0 1.5 2.1 1.2 162 0.6 0.2 1.9 0.9 30.4 74.1 2.0 G11 2517 67.8 1.0 69.4 76 1.1 2.0 1.3 1.8 1.8 1.4 169 0.7 0.6 1.6 0.5 26.7 77.5 2.1 G38 2413 69.0 1.1 70.3 78 1.0 1.9 1.0 1.4 1.9 1.4 175 0.5 0.8 1.4 0.8 30.5 72.8 2.0 G2 2326 62.1 1.2 64.5 69 1.0 2.0 1.3 1.9 1.8 1.3 56 1 1.0 0.5 1.5 1.5 29.3 67.4 2.1 G40 2295 66.5 1.1 68.0 73 1.1 1.9 1.2 1.4 1.9 1.2 169 0.8 0.6 1.4 0.5 33.2 77.8 2.0 Mean 3304 64.7 1.1 66.5 76.9 1.1 1.9 1.3 1.6 1.9 .3 1 171.3 0.8 0.7 1.6 0.8 30.4 78.1 2.0 LSD 608 1.7 1.3 1.7 8 0.1 0.3 0.8 0.3 0.5 0.3 12 .4 0 1.2 0.3 0.7 2.9 5.9 0.3 CV %) 31.8 4.6 194 4.6 18.2 15.2 27.1 103 36 40 35.9 1 12.80.3 302 31.3 169 16.8 13.1 27.3 Min 2295 59.1 0.6 61.1 66.0 1.0 1.4 0.9 1.3 1.5 0 1. 156.0 0.5 0.1 1.3 0.0 25.9 67.4 1.5 Max 4489 69.0 2.9 70.3 89.0 1.2 2.4 2.6 2.1 2.7 7 1. 189.0 1.5 1.8 2.1 1.8 33.7 86.5 2.5 LSD = Least significant difference, CV = coeffitc oief nvariation, Min = minimum, Max = maximum, GYg =ra in yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi l interval (days), DS = days to silking (days), EeHa r= height (cm), EPP = ears per plant (#), LB =f lbelaight disease (1-5), HC = husk cover, GLS = glerayf spot disease (1-5), GT = grain texture (1-5), MSV = maize streak virus dissee (a1-5), PH = plant height (cm R)E, = rotten ears (#), RL = root lodging (#), Ru srtu =st disease (1-5), SL= stem lodging (#), SWT = 100 seed weight (g), SH = shelling percentage,= V vIGig our (1-5), # = number. 96 4.5.14 Pearson’s correlation coefficients betwreaei n ygield and other agronomic traits across optimal environments combined fo1r/ 1220 1and 2012/13 seasons Days to anthesis were positively and significanctolyr related with all traits apart from anthesis-silking interval (Table 4.16). Grain yi ewldas positively and significantly correlated with days to anthesis days (r = 0.1a6y),s dto silking (r = 0.16), plant and ear height (r = 0.16 and r = 0.14, respectively) as l waesl with number of ears per plant (r = 0.80) and shelling percentage (r = 0.91). Ntievgea correlations among other traits were recorded for example anthesis-silking interval wneagsa tively correlated with root lodging (r = -0.10), husk cover (r = -0.18), leaf blight =(r -0.20) and gray leaf spot (r = -0.10) disease as well as with plant vigo=r (-r0 .33). Anthesis-silking interval also correlated positively with traits such as days itlkoi nsg (r = 0.19), plant and ear height (r = 0.27 and 0.25, respectively), MSV (r = 0.0n9d) a100 seed weight (r = 0.18). Days to silking correlated significantly with all traits ceexpt for plant vigor. Ear rots were positively and significantly correlated with stem lodging ( r0 =.30), husk cover (r = 0.20), grain texture (r = 0.20), leaf blight (r = 0.50), graya fl espot (r = 0.40), rust (r = 0.30), MSV (r = 0.60), 100 seed weight (r = 0.30) and plant vigr o=r 0(.30). Plant (r = -0.33), and ear height (r = -0.40), were negatively correlated with plavnigt our. This implies that tall plants (higher values) were associated with good vigou lor waser scores of plant vigour (below 3 in the scale of 1-5) indicates good plant vigoudr ahnigher scores indicate poor vigour and short plants. Shelling percentage furthermore claotrerde positively and significantly with number of ears per plant (r = 0.90), plant heigr h=t 0( .17) and ear height (r = 0.20) (Table 4.16). 97 Table 4.16 Pearson's correlation coefficientsr faoinr gyield and agronomic traits across all opteimnavli ronments for two seasons 2011/12 and 2012/13 GY AD ASI DS PH EH RL SL HC RE GT LB GLS Rust MSV SWT VIG SH AD 0.16** ASI 0.04 0.09 DS 0.16** 1.00** 0.19** PH 0.16** 0.76** 0.27** 0.80** EH 0.14** 0.60** 0.25** 0.60** 0.93** RL 0.03 0.25** -0.10** 0.20** -0.04 -0.10 SL 0.03 0.31** -0.08 0.30** 0.00 -0.04 0.40** HC 0.04 0.29** -0.18** 0.30** 0.10** 0.02 0.20** 02.0** RE 0.03 0.39** -0.03 0.40** 0.06 -0.10 0.06 0.30** 0.20* GT 0.08 0.12** -0.06 0.10** -0.20** -0.20 0.10** 100. ** 0.10** 0.20** LB 0.07 0.60** -0.20** 0.60** 0.20** 0.10 0.40** 04.0** 0.50* 0.50** 0.30** GLS 0.08 0.60** -0.10** 0.60** 0.30** 0.20** 0.20** 0.30** 0.40* 0.40** 0.30** 0.70** Rust 0.07 0.46** -0.06 0.50** 0.32** 0.18 -0.20 03. 0 0.30* 0.30** 0.30** 0.50** 0.60** MSV 0.05 0.45** 0.09** 0.50** 0.25** 0.18 -0.04 00.2** 0.00 0.60** 0.30** 0.50** 0.40** 0.40** SWT 0.08 0.58** 0.18** 0.60** 0.36** 0.30** 0.20** 0.30** 0.00 0.30** 0.40** 0.50** 0.50** 0.30** 0.60** VIG 0.04 0.12** -0.33** 0.09 -0.33** -0.40* 0.40** 0.40** 0.30* 0.40** 0.30** 0.70** 0.50** 0.20** 0.30** 0.20** SH 0.91** 0.18** 0.04 0.20** 0.17** 0.20** 0.03 0.30 0.00 0.03 0.10** 0.08 0.10** 0.08 0.06 0.10** 0.*1* EPP 0.80** 0.16** 0.04 0.20** 0.16** 0.10** 0.03 03. 0.00 0.03 0.10** 0.07 0.08 0.07 0.05 0.08 0.08 .90*0* **P ≤0.01; GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daPyHs) =, plant height (cm )E,H = ear height (cm), RL = root lodging (#), SL = stem lodging (#), HC =k h cuosver, RE = rotten ears (#), GT = grain textu1re-5 (), LB = leaf blight disease (1-5), GLS = graayf lsepot disease (1-5), Rust = rust disease (1-5), MSV = maize streak virus diseas5e), (S1W- T = 100 seed weight (g), VIG = vigour (1-S5H), = shelling percentage, EPP = ears per plan t# ( #=) n, umber. 98 4.5.15 Principal component analysis results, eaiglueensv and eigenvectors for the traits across four optimal environments combinre 2d0 f1o1/12 and 2012/13 seasons PCA generated 16 principal components (Table 4w.1h7i)c h accounted for 100% of variability present in the maize genotypes evaldu.a Ttehe first eight PCs had eigenvalues within a range of 1.0 – 7.1. Their cumulative penrtcaeges accounted for 85.9% of the total variation present among the genotypes. Table 4.17 Eigenvalues and eigenvectors for thites tarcaross four optimal environments combined for 2011/12 and 2012/13 nsse aso PC Eigenvalues As percentages Cumulative percesn tage 1 7.16 39.75 39.75 2 1.62 8.98 48.74 3 1.43 7.97 56.70 4 1.30 7.22 63.93 5 1.00 5.56 69.48 6 0.96 5.31 74.79 7 1.00 5.56 80.35 8 1.00 5.56 85.91 9 0.62 3.45 89.36 10 0.48 2.65 92.01 11 0.43 2.40 94.41 12 0.36 2.00 96.41 13 0.29 1.60 98.01 14 0.18 0.99 99.00 15 0.17 0.92 99.92 16 0.01 0.08 100.00 4.5.16 Clustering of the maize genotypes evaluaat tfeodu r optimal environments combined for the 2011/12 and 2012/13 seasons Forty five maize genotypes were clustered basetdh eoinr means for grain yield. At cut off point 1.0, three main clusters were observed (Fei g4u.2r) with a high cophenetic correlation of rcop = 0.82. Thirteen genotypes from CIMMYT-Zimbabwe wegrreo uped in cluster II with only two from CIMMYT-Colombia. The top perforimng genotypes were grouped in cluster III which comprised of 24 genotypes fromth b Coolombia and Zimbabwe. Cluster I was comprised of six genotypes which were alosom f rboth research centers. 99 Genotypes G11 G37 G7 G5 G34 G38 G40 G2 G20 G36 G43 G32 G45 G28 G30 G44 G12 G31 G10 G35 G22 G26 G18 G6 G33 G27 G8 G 9 G15 G21 G 4 G16 G 42 G 24 G3 G23 G41 G17 G19 G14 G1 2.50 2.08 1.67 1.25 0.83 0.42 0.00 Dissimilarity Figure 4.2 Dendrogram based on Euclidean distandce U aPGMA clustering using morphological data for genotypes at four optimvairl oenments combined for 2011/12 and 2012/13 seasons 100 4.5.17 Combined ANOVA for grain yield and agrono tmraicits for all environments, optimal and low pH for two seasons 2011/12 and/ 1230 12 Mean squares from the combined ANOVA across alli reonvments and two seasons are given in Table 4.18. Genotype mean squares werhel yh sigignificant (P≤0.01) for grain yield, days to anthesis, days to 50 % silking, heeaigr ht, leaf blight disease, husk cover, grain texture, maize streak virus, rust diseasoet, lrodging, 100 seed weight, shelling percentage and plant vigour and significantly≤ 0(.P05) with number of ears per plant. Environment mean squares were highly significarn at lflo the traits measured. Season mean squares were highly significant for days to antsh,e asni thesis-silking interval, days to 50 % silking, number of ears per plant, plant heightt,e rno ears and all the diseases, 100 seed weight as well as shelling percentage. GxE wasl yh isgihgnificant for grain yield, anthesis date, days to 50% silking, number of ears per p, lhaunstk cover, grain texture, rotten ears, 100 seed weight, and plant vigour as well as MSdV raunst. Interaction of GxY was not significant for anthesis-silking interval, numbefr eoars per plant and stem lodging. ExY mean squares were significant for most traits boutt fnor husk cover and rotten ears. Interaction of GxExY was highly significant for ginra yield, anthesis date, days to 50% silking, grain texture, root and stem lodging, lebaligf ht disease and 100 seed weight as well as plant vigor. 4.5.18 Estimated contributions to total sum of rseqsu aacross all environments for two seasons 2011/12 and 2012/13 Contribution to total sum of squares for genotypaes wthe highest for 100 seed weight (4.88%) and MSV (4.85%, Table 4.19). Contributioune dto environment was highest for days to 50% silking (80.45%) and for season theh ehsigt was rust disease (17.0%). GxE contribution was significant for husk cover (26.3. %T)he interaction of GxY made the highest contribution to gray leaf spot (5.7%). Tinhtee raction of ExY made the highest contribution to root lodging (21.9%) and the intcetiroan of GxExY made the highest contribution to shelling percentage (11.03%) (T a4b.1le9). 101 Table 4.18 Mean squares for combined ANOVA for ng ryaiield and agronomic traits for all environmeonptsti,m al and low pH for two seasons 2011/12 and 2012/13 Source Genotypes Environment Year GxE GxY ExY GxExY MSE Df 44 7 1 307 44 6 264 1334 GY 3E+06** 4.5E+08** 1.81E+06 1E+06** 3E+06** 3E+0**7 1E+06** 764200 AD 129.46** 41689.99** 1709.56** 32.2** 59.5** 487.022** 37.4** 13.54 ASI 4.99 712.359** 285.673** 4.24 3.73 191.02** 46.58* 3.99 DS 114.3** 45472.23** 666.96** 33.53** 59.48** 518.846** 36.23** 13.87 EH 518.4** 132261.7** 1176.5* 340.3* 458.9** 157510*.* 182.2 191.3 EPP 0.16* 9.09** 20.16** 0.12** 0.15 4.58** 0.13 102. LB 0.88** 83.57** 12.37** 0.51 1.14** 42.15** 0.83* 0.44 HC 4.12** 192.39** 4.14 4.43** 3.16* 0.20 2.55 2. 04 GLS 1.02 55.86** 9.04** 1.21* 2.86** 37.23** 1.20* 0.99 GT 1.52** 55.77** 0.01 1.51** 1.52** 18.64** 0.78* * 0.65 MSV 0.41** 16.45** 20.68** 0.28** 0.41** 5.35** 0.19 0.20 PH 5820 261853** 61886** 5203 8806** 52208** 6018 3459 RE 1.20 97.71** 14.17** 2.02** 2.32** 4.31 3. 40 010. RL 6.33** 395.1** 9.30 3.94* 8.66** 595.95** 5.71* * 3.17 Rust 0.83** 89.76** 213.61** 0.42** 0.56** 21.43** 0.27 0.30 SL 3.14* 89.014** 5.55 2.07 2.85 91.38** 2.897** 029. SWT 108.32** 5805.43** 740.77** 34.33** 56.64** 2601.86** 36.15** 22.66 SH% 213.40** 14288.60** 3331.90** 154.30 205.40* 5100.30** 152.30 142.20 VIG 1.72** 190.8** 0.24 0.72** 1.23** 19.75** 0.59* 0.43 **P ≤0.01; *P≤0.05; G = Genotype, E = environment, Y = year, M=S mE ean square error, GY = grain yield (kg- 1h),a AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to 50% silking (days), EH = ear heighmt) (, cEPP = ears per plant (#), LB = leaf blight daissee (1-5), HC = husk cover, GLS = gray leaf spsoet adsie (1-5), GT = grain texture (1-5), MSV = maize streak virus diseases (1-5),= P pHla nt height (cm), RE = rotten ears (#), RL = tr ol odging (#), Rust = rust disease (1-5), SL = slotedmging (#), SWT = 100 seed weight (g), SH = shelling percentage (%), VIG =o vuirg (1-5), # = numbe r 102 Table 4.19 Relative percent contribution to toutaml sof squares across two years at eight environs m(oepnttimal and low pH) Source GY AD ASI DS EH EPP LB GLS GT HC MSV PH RE LR Rust SL SWT SH VIG Genotypes 2.55 1.54 1.49 1.27 1.59 2.06 2.73 2.02 .29 3 4.10 4.85 2.07 2.52 2.56 2.90 3.40 4.88 2.21 31 3. Environment 57.80 79.10 33.79 80.35 64.70 18.13 523 5. 15.12 19.25 26.11 22.31 14.81 27.90 21.81 42.7113 .14 35.66 23.59 58.56 Year 0.03 0.46 1.94 0.17 0.08 5.74 0.88 0.41 0.00 .09 0 5.61 0.50 0.67 0.09 16.94 0.14 0.76 0.79 0.01 GxE 6.73 2.68 8.84 2.60 7.30 10.26 9.49 14.39 2 2.7246.32 16.57 12.90 25.38 9.56 8.83 13.46 9.28 1 1.291.73 GxY 2.38 0.71 1.11 0.66 1.41 1.87 3.54 5.68 3.30 14 3. 4.89 3.13 4.85 3.50 1.96 3.08 2.55 2.13 2.35 ExY 3.38 7.92 7.77 7.85 5.50 7.83 8.96 6.72 4.60 00 0. 2.90 2.11 1.57 21.93 3.40 6.75 13.37 1.73 2.60 GxExY 7.24 2.68 8.70 2.41 2.80 9.70 7.74 9.49 8.462 .54 4.46 10.70 1.31 9.24 1.91 9.41 7.99 11.03 3.42 Rep 0.17 0.00 0.00 0.00 0.00 0.20 0.30 0.40 0.00 70 0. 0.30 0.10 0.10 0.10 0.60 0.10 0.50 011 0.00 Residual 19.9 4.90 36.34 4.67 16.61 44.18 30.87 774 5. 38.35 37.02 38.13 53.67 35.73 31.25 20.79 50.5244 .96 47.8 19.99 CV (%) 36.00 5.00 103.10 4.90 22.20 35.1o 34.90 009 4. 54.80 34.40 36.50 48.90 101.00 106.00 36.40 8104 2.18.40 11.90 27.50 Min 0.03 0.004 0.03 0.009 0.003 0.22 0.28 0.41 0.000.005 0.28 0.12 0.07 0.05 0.56 0.09 0.54 0.11 0.01 Max 57.80 79.10 36.30 80.40 64.70 44.20 35.50 4 5.8308.30 37.00 38.10 53.70 35.70 31.30 42.70 50.50 .703 5 47.80 58.60 G = Genotype, E = environment, Y = year, CV = cioceieffnt of variation, Min = minimum, Max = maximuGmY, = grain yield (kg h-a1), AD = days to anthesis (days), ASI = anthesis- silking interval (days), DS = days to silking ()d, aEyHs = ear height (cm), EPP = ears per plant (L#B), = leaf blight disease (1-5), GLS = gray leaft sdpisoease (1-5), GT = grain texture (1-5), HC = husk cover, MSV = maize streak viruse daise (1-5), PH = plant height (c mRE), = rotten ears (#), RL = root lodging (#), Rus rtu =st disease (1-5), SL = stem lodging (#), SWT = 100 seed weight (g), SH = shelling percen, tVaIgGe = vigour (1-5), # = number. 103 4.5.19 Genotypic and phenotypic variance compo,n bernotasd sense heritability and genetic advance estimates across the combinedo nemnveinrts for both 2011/12 and 2012/13 seasons The results in Table 4.20 indicated that phenot yvpairciances were higher than genotypic variances. Grain texture (0.6), days to 50% sil k(i0n.g5), anthesis date (0.5) and plant vigour (0.5) had relatively high broad sense hbeirliittay estimates. The genetic coefficients of variations were lower than phenotypic coeffictsie nof variation (Table 4.21). The expected genetic advance was highest for planth ht efoigllowed by grain texture. Shelling percentage had the lowest genetic advance valuhe aw hitigh phenotypic coefficient of variation. Table 4.20 Genotypic variances, phenotypic varsia anncde heritability estimates across optimal and low pH environments for 201a1n/1d2 2 012/13 seasons Trait σ2 2 2g σ p H b H2b % GY 2500.00 401500.000 0.006 0.623 AD 21.250 39.300 0.541 54.071 ASI 1.200 4.400 0.273 27.273 DS 22.345 40.645 0.550 54.976 EH 53.250 238.850 0.223 22.294 EPP 0.055 0.275 0.200 20.000 LB 0.145 0.735 0.197 19.728 GLS 0.450 2.050 0.220 21.951 GT 0.630 1.120 0.563 56.250 HC 1.485 3.505 0.424 42.368 MSV 0.015 0.185 0.081 8.108 PH 3290.00 13596.000 0.242 24.198 RE 0.035 1.395 0.025 2.509 RL 0.540 2.740 0.197 19.708 Rust 0.040 0.360 0.111 11.111 SL 0.280 2.780 0.101 10.072 SWT 19.815 40.785 0.486 48.584 SH 18.700 157.900 0.118 11.843 VIG 0.620 1.180 0.525 52.542 σ2g = genotypic varianceσ,2 p = phenotypic variance, 2bH = broad sense heritability,2 bH % = broad sense heritability as percentage, GY = grain yield (kg-1) h, Aa D = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (days), EH= ear he(icgmht) , EPP = ears per plant (#), LB = leaf blight adsisee (1-5), GLS = gray leaf spot disease (1-5), GT = grain teex t(u1r-5), HC = husk cover, MSV = maize streak vidrui se ase (1-5), PH = plant height (cm R),E = rotten ears (#), RL = root lodging (#), Rustu =s tr disease (1-5), SL= stem lodging (#), SWT = 100 seed weight (g), SH = shge pllienrcentage, VIG = vigour (1-5), # = number. 104 Table 4.21 Genotypic coefficient of variation, pohtyepnic coefficient of variation and genetic advance across all eight environmentws ofo yre tars Trait GCV% PCV% GA GA (% of mean) GY 10.1 128.5 0.3 10.6 AD 5.4 7.3 2.1 2.9 ASI 7.9 15.1 1.2 63.6 DS 5.4 7.3 2.1 2.8 EH 9.2 19.6 1.1 1.7 EPP 2.4 5.3 0.7 70.8 LB 2.8 6.2 0.7 36.8 GLS 5.0 10.6 0.9 49.6 GT 5.2 6.9 2.2 92.6 MSV 1.1 3.9 0.1 9.0 PH 46.9 95.4 1.7 108.0 RE 1.9 11.7 0.1 5.6 RL 5.7 12.8 0.5 31.2 Rust 1.6 4.9 0.2 11.9 SL 5.3 16.6 0.3 26.5 SWT 8.8 12.6 2.5 9.8 SH 5.0 14.6 0.9 1.2 VIG 5.1 7.0 2.0 85.2 GCV = genotypic coefficient of variation, PCV = npohteypic coefficient of variation, GA = genetic nacdev,a GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (days), EH = ear height (cm), EPP = ears per pla#n),t L(B = leaf blight disease (1-5), GLS = gray lepaoft sdisease (1-5), GT = grain texture (1-5), MSV = maize strevaikr us (1-5), PH = plant height (cm R),E = rotten ears (#), RL = root lodging (#), Rust = rust disease (1-5), S Ls te=m lodging (#), SWT = 100 seed weight (g), SH = shelling percentage, VIG = vigour (1-5), #u =m nber. 4.5.20 Mean performance for grain yield and othraeirt st across all environments for 2011/12 and 2012/13 seasons The grain yield trial mean was 2.4 t- 1h (aTable 4.22) and the top performing genotype was G20. None of the top ten genotypes were prolifitch wai mean number of ears per plant of 1. The genotype most susceptible to gray leaf dspisoeta se was again G 41. The tallest genotype was G7 (213 cm), although it ranked whiteh btottom ten genotypes. Genotype G27 recorded the highest 100 seed weight (27 glo) wfoeld by G35 (26 g). In terms of shelling percentage, genotype G30 (74%) recorde dh itghhest. 105 Table 4.22 Mean performance combined across twros yaenad across optimal and low pH environments0 f1o1r/ 122 and 2012/13 seasons Entry GY AD ASI DS EH EPP LB HC GLS GT MSV PH RE RL Rust SL SWT SH VIG Top 10 G20 2977 74.4 1.8 76.7 63.8 1.0 1.7 1.5 1.92.3 1.2 146 1.0 1.6 1.3 0.9 27.5 77.3 1.9 Genotypes G27 2850 75.3 1.8 77.3 67.1 1.0 1.7 1.3 .7 1 2.6 1.2 153 1.0 1.2 1.4 1.1 29.3 74.2 2.1 G8 2727 74.3 1.8 76.4 65.6 1.0 2.0 1.4 2.0 2.2 1.2 153 0.8 1.6 1.6 1.4 26.3 73.0 2.2 G33 2725 72.1 2.1 74.9 60.9 1.0 1.7 1.3 1.7 2.4 3 1. 145 0.8 1.9 1.5 1.2 24.8 76.0 2.2 G5 2719 76.7 1.5 78.5 65.4 1.0 1.9 2.0 1.7 2.3 1.1 156 1.0 1.3 1.4 1.2 27.6 75.4 2.0 G30 2704 73.6 2.0 76.0 63.5 1.0 1.8 1.8 1.9 2.2 3 1. 147 1.0 1.6 1.5 0.6 27.2 76.3 2.1 G39 2680 72.8 2.0 75.4 59.9 1.0 1.9 1.2 1.8 2.3 2 1. 145 1.1 2.4 1.5 1.2 26.7 73.6 2.5 G45 2678 74.8 1.9 77.1 67.4 0.9 1.7 1.5 1.6 2.8 4 1. 157 0.9 1.2 1.5 0.8 28.7 72.5 2.1 G22 2663 71.8 2.0 74.3 62.3 1.0 1.9 1.6 1.7 2.3 3 1. 146 1.0 1.8 1.5 1.0 26.1 75.1 2.6 G28 2635 72.9 1.7 75.0 62.1 1.0 2.0 1.4 1.9 2.7 2 1. 147 0.8 1.8 1.6 0.9 25.2 75.0 2.3 Bottom 10 G16 2275 71.9 2.0 73.9 60.0 0.9 1.7 2.0 .9 1 2.3 1.2 143 0.9 1.2 1.6 1.2 27.8 74.2 2.3 Genotypes G41 2264 69.8 1.7 72.1 54.4 0.9 1.7 1.5 .1 2 2.1 1.1 137 0.9 1.4 1.6 0.8 23.9 75.1 2.5 G1 2259 74.9 3.1 77.1 63.9 0.9 2.0 1.2 1.9 2.1 1.2 151 1.1 1.6 1.5 1.0 26.6 72.2 2.5 G42 2185 72.7 1.9 75.1 63.7 0.9 2.0 1.3 1.6 2.1 2 1. 152 1.1 1.8 1.5 0.9 26.0 73.3 2.7 G7 2095 72.8 2.3 75.1 61.7 0.9 1.9 1.0 1.6 2.6 1.3 213 1.1 1.4 1.3 0.8 26.7 76.3 2.4 G11 2009 74.7 1.6 76.6 56.2 0.9 2.0 1.4 2.0 2.3 3 1. 140 0.8 1.5 1.5 1.2 22.4 74.6 2.4 G37 2001 73.6 1.7 75.9 56.3 0.9 1.7 1.7 1.7 2.6 2 1. 142 0.9 1.1 1.7 1.1 24.6 72.3 2.3 G40 1884 75.2 1.8 77.2 60.8 0.9 1.8 1.5 1.6 2.3 2 1. 149 0.9 1.7 1.4 0.7 26.5 73.2 2.2 G38 1883 77.1 1.3 78.5 62.8 0.9 1.8 1.3 1.7 2.3 4 1. 147 1.1 1.0 1.3 0.9 25.8 71.1 2.3 G2 1804 71.8 1.7 74.3 57.6 0.9 2.1 1.4 1.9 2.1 1.2 137 0.9 2.4 1.4 1.5 24.8 66.4 2.6 Mean 2401 73.7 1.9 75.9 61.8 1.0 1.9 1.5 1.8 2.3 2 1. 150 1.0 1.6 1.5 1.0 26.2 73.8 2.3 LSD 506 2.4 1.01 2.4 3.9 0.3 0.44 0.8 0.7 0.56 0.3658 0.68 0.94 0.32 0.9 2.6 7.5 0.4 CV 36 5 103 4.9 22.2 35 34.9 94 55 34.4 36.5 49 101106 36.4 143 18 11.9 28 Min 1804 69.8 1.3 72.1 54.4 0.9 1.7 1.0 1.6 2.1 1.1 137 0.8 1.0 1.3 0.6 22.4 66.4 1.9 Max 2977 77.1 3.1 78.5 67.4 1.0 2.1 2.0 2.1 2.8 1.4 213 1.1 2.4 1.7 1.5 29.3 77.3 2.7 LSD = Least significant difference, CV = coeffitc oief nvariation, Min = minimum, Max = maximum, GYg =ra in yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi l interval (days), DS = days to silking (days), EeHa r= height (cm), EPP = ears per plant (#), LB =f lbelaight disease (1-5), HC = husk cover, GLS = glerayf spot disease (1-5), GT = grain texture (1-5), MSV = maize streak virus dissee a(1-5), PH = plant height (cm R)E, = rotten ears (#), RL = root lodging (#), Ru srtu =st disease (1-5), SL = stem lodging (#), SWT = 100 seed weight (g), SH = shelling percentage,= V vIGig our (1-5), # = number. 106 Table 4.23 Pearson’s correlation coefficientsr faoinr gyield and agronomic traits across optimall oanwd p H environments for 2011/12 and 2012/13 seasons GY AD ASI DS PH EH RL SL HC RE GT LB GLS RUST MSV SWT VIG SH AD 0.12** ASI 0.23** 0.09** DS 0.12** 0.99** 0.09** PH 0.05 0.14** -0.01 0.11** EH 0.06** 0.02 0.01 0.01 0.62** RL 0.03 0.20** 0.01 0.20** -0.01 0.02 SL 0.03 0.25** 0.01 0.24** -0.01 -0.1** 0.21** HC 0.01 0.18** 0.00 0.15** 0.11** 0.05 0.03 0.09** RE -0.02 0.35** -0.03 0.34** -0.06** -0.2** -0.02 .019** 0.20** GT 0.06 0.09 0.01 0.09 -0.06 -0.14 0.05 0.03 -0.020 .10 LB 0.04 0.40** 0.01 0.40** 0.06 -0.01 0.28** 0.29 ** 0.25** 0.34** 0.3** GLS 0.03 0.38** 0.00 0.37** 0.08** 0.00 0.18** 0.1**4 0.15** 0.26** 0.27** 0.58** RUST 0.01 0.22** 0.00 0.20** 0.23** 0.18** -0.2** .007** 0.30** 0.27** 0.28** 0.51** 0.45** MSV 0.02 0.35** -0.02 0.33** 0.08** 0.03 0.00 0.2*2 * 0.08** 0.41** 0.26** 0.58** 0.39** 0.48** SWT 0.07** 0.45** 0.02 0.44** 0.21** 0.25** 0.30** 0.18** -0.01 0.16** 0.25** 0.32** 0.30** 0.17** 0.4** VIG 0.04 0.20** 0.01 0.19** -0.1** -0.2** 0.38** 02. 4** 0.12** 0.25** 0.22** 0.48** 0.34** 0.17** 0.18** 0.3** SH 0.55** 0.15** 0.19** 0.15** 0.09** 0.11** 0.05 0.04 0.05 0.04 0.12** 0.09** 0.08** 0.08** 0.07** 01.2** 0.09** EPP 0.46** 0.24** 0.19** 0.24** 0.10** 0.10** 0.05 0.04 0.05 0.04 0.13** 0.09** 0.08** 0.08** 0.07** .012** 0.10** 0.6** **P ≤0.01; GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daPyHs) =, plant height (cm )E,H= ear height (cm), RL = root lodging (#), SL= stem lodging (#), HCu =s kh cover, RE = rotten ears (#), GT = grain tex t(u1r-e5), LB = leaf blight disease (1-5), GLS = glreaayf spot disease (1-5), Rust = rust disease (1-5), MSV = maize streak virus dies e(1a-s5), SWT = 100 seed weight (g), VIG = vigou-r5 )(,1 SH = shelling percentage, EPP = ears per p (l#a)n, t# = number. 107 4.5.21 Pearson’s correlation coefficients betwreaei n ygield and other agronomic traits across optimal and low pH environments cnoemdb fior 2011/12 and 2012/13 seasons Grain yield was positively and significantly corarteeld with number of ears per plant (r = 0.46), anthesis date (r = 0.12), anthesisi-nsgil kinterval (r = 0.23), day to 50% silking (r = 0.12), ear height (r = 0.06), 100 seed wei(gr h=t 0.07) and shelling percentage (r = 0.55) (Table 4.23). Correlation among othearit str indicated that shelling percentage was positively and significantly correlated witht haensis date (r = 0.15), anthesis-silking interval (r = 0.19), days to 50% silking (r = 0. 1a5n)d number of ears per plant (r = 0.06) as well as plant height and ear height (r = 0.0d9 0a.n11). Ear rots were positively and significantly correlated with traits such as daoy sa nt thesis (r = 0.35), day to 50% silking (r = 0.34), stem lodging (r = 0.19), husk cover (r. 2=0 0), leaf blight (r = 0.34), gray leaf spot (r = 0.26), rust (r = 0.27), MSV (r = 0.41). Eari ghhet (r =-0.2) was negatively correlated with plant vigour and seed size (100 seed weighats) wpositively correlated with grain texture (r = 0.25) and plant vigour (r = 0.3). 4.5.22 Principal component analysis results, eaiglueensv and eigenvectors for the traits across all environments combined for 20 1a1n/d1 22012/13 seasons Results indicated that eight principal componenetsre w generated across all environments and these accounted for 100% variability prese ntht ein maize genotypes evaluated (Table 4.24). The first five PC had higher Eigenvalues tahnedir cumulative percentages accounted for 71.1% of the total variation present amongg tehneo types. 108 Table 4.24 Eigenvalues, percentages and cumulpaetirvce ntages for the measured and derived data across four low pH soil environtsm ceonmbined for 2011/12 and 2012/13 seasons PC Eigenvalues As percentages Cumulative percesn tage 1 2.2 18.5 18.5 2 1.8 15.7 34.2 3 1.5 13.1 47.4 4 1.5 12.5 59.8 5 1.3 11.3 71.1 6 1.2 10.0 81.1 7 1.1 9.7 90.8 8 1.1 9.2 100.0 Table 4.25 Phenotypic and genotypic variancesr afoinr gyield and other traits at four optimal environments Trait σ2 2 2 g σ p H b AD 2.26 4.34 0.52 DS 2.19 4.34 0.51 EPP 3E-04 0.001 0.27 GY 669.20 2E+05 0.01 σ2g = genetic advanc 2, σe p = phenotypic varianc, He2b =broad sense heritability, AD = days to anth DesSi s=, days to silking (days), EPP = ears per plant (#), GY a=i ng ryield (kg h-a1), # = number. 4.5.23 Clustering of the maize genotypes evaluaat tfeodu r low pH and four optimal environments combined for the 2011/12 and 2012e/a1s3o sns The dendrogram was constructed using the UPGMAt ecrl uasnalysis method based on morpho-agronomic data across optimal and low pHir oenmv ent Figure 4.3. Cluster two genotypes comprised of well adapted released OP4V4s (GZM523) and G43 (ZM309) alongside some four of the top ten based on meaain gyireld across optimal and low pH environment (Table 4.22) G30, G33, G8 and G5. Tho Cugluster three was larger than cluster one, the two did not show any clear pa totef rcnlustering. Both groups of genotypes from CIMMYT-Colombia and CIMMYT-Zimbabwe had equcahl ances to be grouped in the two clusters. 109 Genotypes G38 G11 G27G37 i G45 G30 G35 G 36 G25 G13 G33 G10 G28 G8 ii G44 G43 G24 G18 G26 G5 G22 G34 G3 G9 G2 G37 G7 G41 G16 G14 G42 iii G17 G12 G 34 G6 G 19 G4 G 23 G31 G 1 2.50 2.08 1.67 1.25 0.83 0.42 0.00 Dissimilarity Figure 4.3 Dendrogra m based o n Euclidea n distandce U aP 1G10M A clus tering usin g morphological data fonro gtyepes at four low pH and four optimal environments combined for 2011/122 a0n1d2 /13 seaso ns 4.6 Discussion Low pH soil is one of the most constraining eda pfhaictors contributing to low crop production. Even liming is known to be inefficieanst it is restricted to the top soil layer while the sub-soil surface layers with toxic Al remainid a (cCustódioe t al., 2002). In the present study, soil analysis from the 30 cm soil profiled iicnated that the sites had different soil pH levels. Lunyangwa low pH site had the most acidoicils sas compared to Tsangano and Bembeke low pH sites. Bvumbwe was classified asin hga mv oderately acidic soils based on the Soil Test Interpretation Guide (Horneectk a l., 2011). In terms of inherent soil fertility with due considerations to resource poor farmers, ea lsl itthes had a low N concentration in the top soil. The across site analysis for low pH sitesti,m oapl sites and combined across environments and seasons indicated highly significant effectrs m fo st of the traits (Tables 4.3, 4.11 and 4.18). This was an indication that genetic variatyb ielxisted and suggested the possibility of selection and further improvement. The across sAitNesOVA for low pH environments revealed that sites were significantly differentd atnhis confirmed soil analysis results. Similarly the ANOVA for optimal sites were also nsifgicantly different owing to the fact that Chitedze and Meru were in the mid-altitude ecolwoghyil e Baka and Chitala were in the low- land ecology. Seasons were also significantly rdeifnfet for a number of traits, suggesting that selection from one season’s data may not be rel ibaubtl across season data could provide reliable information about the genotypes. Partitniogn of variation into four sources of interactions followed the same trend such thaet tfhfeec ts of GxE, GxY, ExY, and GxExY were also significant for a number of traits. Different sources of variation made different coibnuttrions to total variation with respect to various traits (Tables 4.4, 4.12 and 4.29). On aagve,r environments made the highest contribution to total variation in all tested enovnirments. Vianae t al. (2009) indicated that spatial heterogeneity in experimental areas ism am con fact and is related to the processes of soil genesis. 111 In general, the effects of low pH soil contributteod reduction in grain yields and yield components under low pH (Tables 4.5 and 4.7). Tohmeb cined mean reduction was 69.9% and this is consistent with other findings (Welckeetr al., 2005). Plant height reduction, reduced number of ears per plant and shelling percentadguec rtieon under low pH soil conditions could be due to the indirect effect of the impaired neunttr iuptake by maize plants as a result of inhibition of root development. This confirms retsu ol f Duque-Vargae t al. (1994) who found that root inhibition leads to low water and nutrt ieunptake and low maize yields. The mean grain yield in combined analysis for low pH siteass w 1.5 t h-a1 (Table 4.8). This is just slightly above the average maize yields for the country t( h1a.3-1) (MOA, 1994). The mean grain yield trial in the combined ANOVA for optimal conditionwsa s 3.3 t h-a1 (Table 4.15 )and for the combined environments was 2.4 t-1 h (aTable 4.22). Comparison of the glasshouse hydnriocp o trial and field trial results indicated that of tthoep 10 yielding genotypes, SYN DTE-STY-W- B performed well and ranked first in terms of RTitih w a NSRL of 2.5 cm and this was followed by VPO717 with RTi (1.0) and NSRL (1.7 cm). Thesees urlts also confirm that the best performer in the field may not be the most tole rtaon tlow pH stress because low pH is considered a complex stress which is associatehd d wisiteases and other stresses. The partitioning of variance into its componentlso wasl plant breeders to estimate the relative importance of the various determinants of the phtyepneo, in particular the role of heredity versus environment (Duvick, 1986; Volenetc al., 2002). In the present study (Tables 4.6, 4.13 and 4.20), the obtained phenotypic varianδc2eps) (were higher than genotypic variances (δ2g) at all the environments with low pH sites recinogrd relatively lower values for both variance components. According to Dabholkar (1992) heritability of a crahcater is classified as low (5-10%), medium (10-30%) and high (30-60%). In the present stuedsyu, lrts for low pH environments combined across seasons recorded high heritability valure sa nfothesis date, days to 50% silking, husk cover, grain texture, 100 seed weight and planot uvri.g At optimal environments 18 traits had high heritability values. Only root lodging (1.97 %h)ad low value while anthesis-silking interval (18.82) and husk cover (13.98%) had med hiuemritability values. The results for 112 optimal environments were consistent with what wreapso rted by Aminu and Izge (2012) and Bello et al. (2012). Genetic advance shows the degree of gain thate c aonb tbained in a character under a particular selection pressure. High genetic advance couplethd hwigi h heritability estimates offers the most suitable condition for selection (Beellto al., 2012). Results for the optimal environment (Table 4.14) indicated higher genetic advance ptioatle fnor grain yield and plant height and this was consistent with what was reported by Vsathshaie t al. (2013). Phenotypic correlation is the association betweweon c tharacters that can be directly observed, or can be determined from measurements of two cthearrsa in a number of individuals of the population (Falconer and Mackay, 1996). In the epnret sstudy, correlation analysis was performed to check if grain yield was associatetdh wsoi me yield and yield components under low pH stress, optimal and across environments le(T a4b.9; Table 4.17; Table 4.23). Significant and positive correlation between gryaieinld with both number of ears per plant and shelling percentage was observed in all combineadly asens. In the present study low values of significant correlation coefficients were recordfoerd certain traits as was expected due to the large dataset used, but only values of 0.2 ande rla wrgere discussed in order to emphasize the most important correlations. In this study low pHtre ss contributed to lower values of significant correlation coefficients due to thee stsr effect at phenotypic level. Positive and significant correlation was observed for 100 seedig wht under low pH environment only and not in the combined environment (Table 4.23). Tehseu lrts were consistent with other reports (Alvi et al., 2003, Sofi and Rather, 2007; Sumathi aMnudralidharen, 2010). Anthesis date was positively and significantly ceolartred with grain yield at low pH and optimal environments except when combined acrosths ebnovironments. This was probably due to differences in ecology for the low-land ompatli conditions. Low altitudes are associated with high temperatures and short seasons suchla tthea mt aturing genotypes tend to perform poorly because they don’t have adequate time tuom auclcate adequate carbohydrates. 113 Under low pH conditions, higher number of days ntoth aesis date were recorded as compared to optimal environments (mean AD value of 83 fowr lpoH versus 65 for optimal). The AD results suggest that selection for low pH shoul d obnee under low pH conditions because what could be classified as early under low pH may neo tt oblerant to other stresses like fungal diseases due to prolonged humid conditions ande lro nsegasons in low pH environments. Reports from other researchers (Brun and Dudle8y9, ;1 B9yrne et al., 1995; Bänzigeer t al., 1997) indicated that selection under stress is meoffreective in maize, than indirect selection under optimal conditions. Negative correlation was observed for grain yienld aear height across low pH and combined low pH and optimal environments and this was cotennstis with other reports (Sćrkeov et al., 2010). Yield was also negatively correlated witahn pt lvigour due to the fact that a higher score in this study for vigour refers to poor vigour hen tche reverse should be true that grain yield was positively correlated with plant vigour. Theo t ware measured in different direction such that a score of one is the best for plant vigou trh ien scale of 1-5 while for grain yield the highest yielding genotype should record the hsigt hneumber of weight units. It implies that an increase in plant vigour is associated withn acnre iase in yield and a decrease in vigour is associated with decrease in yield. Stem lodging a wlsaos positively and significantly correlated with grain yield which was probably due to the irnedcit effect of low pH on the stalks of maize plants. According to Kochiaent al. (2005) Al rapidly and effectively inhibits the liunxf of Ca2+ into cells by modulating the activity of transporsrt eand by changing the membrane potential. Stalks of maize plants under low pH are likely eto w beak due to low levels of Ca which causes low rigidity of cell walls and may not be able tuop sport the weight of maize ears; hence higher yields were associated or correlated with stem inlogd gunder low pH soil environments. Clustering of maize genotypes based on morpholol gdiactaa across low pH environments showed that maize genotypes had similar chancbeesi nogf grouped in any of the four clusters. The top five genotypes in terms of mean grain y pieelrdformance were comprised of two from CIMMYT-Colombia and three from CIMMYT-Zimbabwe. Otnh e other hand when the genotypes were clustered using across optimal oawn dp Hl environment morphological data, the well adapted released OPVs G44 (ZM523) and (GZM433 09) alongside some four of the top ten based on mean grain yield across optimda l oawn pH environment G30, G33, G8 and 114 G5. This suggested that the four entries and tshte i nre this cluster are likely to be adapted to Malawi like the released ZM523 and ZM309. The tweor ew the first drought tolerant OPVs to be released in the country. 4.7 Conclusions and recommendations Low pH soil is one of the abiotic factors contriibnugt to low yields in Malawi. In this study it contributed to reduction in grain yields and yiecoldm ponents for the maize genotypes which were evaluated. The mean reduction in yield anldd ycioemponents of 69.9% due to impaired nutrient uptake confirmed the various reports oen e tfhfect of low pH stress on yield of maize. In this study, genotypes LPHpop16, LPHpop3, VPO 7V3P9O, 5173 and Low N Pool C3-B had relatively better performance under low pH soil dciotinons. It is interesting to note that among the top 10 genotypes it was SYN DTE –STY-W-B thaant kred first in terms of RTi with a NSRL of 2.5 cm in the glasshouse hydroponic expenritm and this was followed by VPO717 with RTi (1.0) and NSRL (1.7 cm). These OPVs wiell cbrossed to disease tolerant lines to generate source populations for inbred lines etxiotrna cfor use in breeding programme for low pH tolerance. Phenotypic traits associated with grain yield, s ausc hplant vigour, 100 seed weight, shelling percentage, number of ears per plant, ear heigdh pt lan t height can be used alongside grain yield when selecting germplasm for tolerance to lpoHw stress. Traits associated with reproduction such as anthesis date and days to s5il0k%ing tend to be influenced by lower temperatures under low pH environments such thralyt meaaturing genotypes mature late and selection may not be effective when done acrosism oapl tand low pH environments. In other words these results suggested that selection wfo rp Hlo tolerance could be effectively carried out in situ and not indirectly under optimal environments. rTeh ies need to establish the cut off point for significant correlation coefficients uenrd stress as the relationship is affected at phenotypic level as opposed to optimal environm ent. 115 4.8 References Agrobase. 2010. Agronomix, Generation II. Agrono msioxftware, Inc. 71 Waterloo St. Winnipeg, Manitoba, R3N054, Canada. Alvi, M.B., M. Rafique, M.S. Tariq, A. Hussain, TM.a hmood and M. Sarwar. 2003. Character association and path coefficient analysis of gyraieinld and yield components maizZee (a maysL .). Pakistan Journal of Biological Sciences 6: -1368. Aminu, D. and A.U. Izge., 2012. Heritability and rCreolation Estimates in MaizeZ e( a may s L.) Under Drought Conditions in Northern Guinea aSnuddan Savannas of Nigeria. World Journal of Agricultural Sciences 8: 598-602. Bänziger, M., F.J. Betran and H.R. Lafitte.1997.f icEiefncy of high nitrogen selection environments for improving maize for low-nitrogeanr gtet environments. Crop Science 37: 1103–1109. Bello, O.B., S.A. Ige, M.A. Azeez, M.S. Afolabi, YS. .Abdulmaliq and J. Mahamood. 2012. Heritability and Genetic Advance for Grain Yield da nits Component Characters in Maize (Zea MaysL .). International Journal of Plant Research 2:- 13485 Brun, E.L. and J.W. Dudley. 1989. Nitrogen respo inns ethe USA and Argentina of corn populations with different proportion of flint andde nt germplasm. Crop Science 29: 565- 569. Byrne, P.F., J. Bolaños, G.O. Edmeades and D.Lo. nE. a1t995. Gains from selection under drought versus multilocation testing in relatedp ticroal maize populations. Crop Science 35: 63-69. Custódio, C.C., D.C. Bomfim, S.M. Saturnino and .N M.Bachado Neto. 2002. Estresse por alumínio e por acidez em cultivares de soja. Sicai eAngtricola 59: 145-153. Dabholkar, A.R. 1992. Elements of Biometrical Geicnse. tConcept Publ. Camp., New Delhi, India. Duque-Vargas, J., S. Pandey, G. Granados, H. Coesb aanlld E.R. Knapp. 1994. Inheritance of tolerance to soil acidity in tropical maize. Crocpi eSnce 34: 50-54. Duvick, D.N. 1986. Plant breeding: Past achievems eanntd expectations for the future. Economic Botany 40: 289-297. 116 Falconer, D.S. and T.F.C. Mackay. 1996. Introdunc ttio Quantitative Genetics th(e4dn) Longman Group Limited. England. FAOSTAT. 2013. Statistical Database of the Food aAngdriculture of the United Nations.http://www.fao.org GenStat. 2013. Introduction to GenStat for Windo 1w6s.1.th edition. VSN International, Hemel Hempstead, Hertfordshire HPI. IES, UK. Hallauer, A.R. and F.J.B. Miranda. 1988 Quantitea tGivenetics in Maize Breeding. Iowa State Univ. Press, Ames. Hintze, J.L. 2007. Number Cruncher Statistical eSmys (tNCSS). Kaysville Utah. USA. Horneck, D.A., D.M. Sulivan, J.S. Owen and J.M. tH., a2r011. Soil Test Interpretation Guide. Oregon State University Extension Service. Kochian, L.V., M.A. Piñeros and O.A. Hoekenga. 2.0 0T5he physiology, genetics and molecular biology of plant aluminium resistance atonxdicity. Plant and Soil 274: 175- 195 Kumwenda, I. and R. Kachule. 2003. Report on Stoalkdehrs’ Meeting on Farmer Organisations in Malawi. Kalikuti Hotel, Lilongwe1,8 th -19th June 2003. Mehlich, A. 1984. Mehlich III soil test extracta nat :modification of the Mehlich II extractant. Communications in Soil Science and Plant Analys5i:s 1 1409-1416. MOA- Ministry of Agriculture. 1994. Guide to Agricltu re Production. Ministry of Agriculture, Lilongwe. Malawi MOA - Ministry of Agriculture. 2003. National Felritzier Strategy, Lilongwe. Malawi. MOA- Ministry of Agriculture, 2005. Crop Estimatefosr 2004/5. Lilongwe. Malawi. Munthali, M.W. 2007. Integrated Soil Fertility Magneament Technologies: A Counteract to Existing Milestone in Obtaining Achievable Economal icCrop Yields in Cultivated Lands of Poor Smallholder Farmers in Malawi. In:v Aandces in intergrated soil Fertility Management in Sub-Saharan Africa: Challenges anpdo rotupnities, pp. 531-536. Sofi, P. and A. G. Rather. 2007: Studies on ge nveatriicability, correlation and path analysis in maize (Zea maysL .). Genetic Cooperation Newsletter. 81: 26-27 Srećkov, Z., J. Boćanski, A. Nastasć,i I. Đalović and M. Vukosavljev. 2010. Correlation and path coefficient analysis of morphological traitfs moaize (Zea maysL .). Research Journal of Agricultural Science 42: 292-296. 117 Sumathi, P. and V. Muralidharan. 20. 1A0nalysis of genetic variability, association andth p a analysis in the hybrids of sesamSee s( amum indicu mL). Tropical Agricultural Research and Extension 3: 63-67 Sumathi, P., Nirmalakumari and K. Mohanraj. 2005e. nGetic variability and traits interrelationship studies in industrially utilizeodil rich CYMMIT lines of maize Z( ea maysL ). Madras Agric. J. 92: 612-617. Vashistha, A., N. N. Dixit, S. K. Sharma and S. Mkearr. 2013. Studies on heritability and genetic advance estimates in maize genotypes. iBeniocsec Discover y4: 165-168. Viana, J.M.S., V.R. Faria and A.C. Silva. 2009. sB iina the prediction of genetic gain due to mass and half-sib selection in random mating potipounlsa. Genetics and Molecular Biology 32: 497-506. Volenec, J.J., S.M. Cunningham, D.M. Haagenson, .W B.eKrg, B.C. Joern and D.W. Wiersma. 2002. Physiological genetics of alfalfa improvem: ePnat st failures, future prospects. Field Crops Research 75: 97-110. Welcker, C., C. The, B. Andreau, C. De Leon, S.P Na.r entoni, J. Bernal, J. Felicite, C. Zonkeng, F. Salazar, L. Narro, A. Charcosset an dJ .W H. orst. 2005. Heterosis and Combining Ability for Maize Adaptation to TropicAalc id Soils: Implications for Future Breeding Strategies. Crop Science 45: 2405-2413. Zambezi, B.T., J.D.T. Kumwenda and R.B. Jones. .1 9C9lo3sing the maize yield gap in Malawi. In: D.C. Munthali, J.D.T. Kumwenda and Fi.s yKombe (Eds.) Proceedings of the Conference on Agricultural Research for Devmeleonpt. Lilongwe and Zomba, Malawi: Department of Agricultural Research and nCcheallor College. 118 CHAPTER 5 Genotype x environment interactions and stabinliatyly asis for tropical and sub- tropical maize genotypes in Malawi 5.1 Abstract Maize, which is the staple food crop in Malawi,w isid ely cultivated in both marginal and favourable arable land, resulting in low yieldss oinm e environments with a high GEI. In this study 45 maize genotypes were evaluated acrosst loeicgahtions for two years using an (0.1) alpha lattice design with three replications. Thbeje cotive was to study the GEI and stability of the tropical and sub-tropical maize genotypes.h Ien At MMI biplot, genotypes LPHpop21, VPO52, VPO72, VPO744 and VPO96 were identifiedt absl es. VPO97 was identified as the most unstable genotype. Chitala optimal site waesn tidfied as the most discriminating environment for the genotypes, while Chitedze oaplt imwas identified as a suitable environment. Clustering of genotypes at cut offn pt o1i.0 indicated a cophenetic correlation of rcop = 0.82 with four clusters and the results were sirm toila that of AMMI such that cluster III was comprised of genotypes that were stable, e xLcPeHptpop21 which was in its own cluster and was considered to be the most stable and ebrefostr mper in terms of grain yield. Similarly clustering of environments at cut off point 1.0 icinadted a cophenetic correlation ocofp r = 0.84 with three clusters that grouped the lowest pHs siinte cluster I and the second most low pH sites in cluster II. All optimal sites were group iend the third cluster. Clustering identified the low nitrogen site as similar to the lowest pH s iitne sterms of its environmental mean. The stable genotypes identified in this study will bsee du as base populations in the breeding programmes to generate new inbred lines. 5.2 Introduction Maize is an important staple food and world-wide it isw iad ely grown cereal crop ranging from 58o north to 40o south with respect to latitude, from sea leve3l 8to0 8 masl in terms of 119 altitude and under 25.4 to 1016 cm annual rain(Bfarlel wbaker, 1985; Hallauer and Miranda, 1988). According to “The Guide to Agricultural Pruocdtion for Malawi” (MOA, 1994), there are two agro-ecologies for maize cultivation in cthoeuntry namely: low-land or low-altitude (<600 masl) and mid-altitude areas. The low-alteit uadrea mostly covers the lakeshore districts and the Shire Valley Districts. This region ha so iwtsn challenges which include prolonged dry spells, early cessation of rains, high tempera tuarneds floods. The mid-altitude areas is the main ecology in terms of size and it is 600 - 13m0a0s l. This is where most of the maize is cultivated and the major environmental challengnecslu ide fungal diseases, low pH, low N and witch weed, which is associated with low-soil fleitryt.i The challenge is to develop varieties of maize wai thhigh grain yield and stable performance in the low pH areas. Some genotypes tend to ex ghoiboidt performance in some environments but not in others (Ramagosa and Fox, 1993). Thiec bcasuse of differences in yield stability between genotypes is the wide occurrence of GEeI.s eT hinteractions can be partly understood as a result of different genotypes responding rdeifnftely to different environmental stresses such as low pH, diseases and other factors. Plraenetd bers endeavour to develop improved genotypes that are superior not only in grain y bieuldt also in a number of other agronomic and quality characteristics over a relatively wide rea ngof environmental conditions. This interaction is important to geneticists and plarnet ebders because of the magnitude of the interaction components which provides information ccerning the adaptation of a given crop variety (Myers, 2004). GEI may alter the performance or development ofr oap cvariety, thus the extent of the environmental effect on a trait determines the irmtapnoce of replicating in time and space such as testing over years and locations. The multiilonca et valuation, however, results in GEIs which are difficult to interpret by plant breedearnsd agronomists and this often reduces the efficiency in selecting the best genotypes (Anniiacrcicho and Perenzin, 1994). The presence of GEI may mean that a superior variety in one tliocna is not necessarily the best in another environment. Kange t al. (1991) suggested that selection based on yielyd monaly not always be adequate when GEI is significant. The analyfs GisE oI is regarded as an important strategy used by plant breeders to evaluate crop varieotire asd faptation and also for making selections 120 for parents for base populations (Aienta a l., 2007). The development of new crop varieties is expensive and time consuming because of the prorec eindvuolved in making sure a stable variety is identified. According to Myers (2004) stability refers to thhea cracter of a crop variety that withstands fluctuations of environments. The stability anasly sfoi r the interpretation of GEI was first proposed by Yates and Cochran in 1938. Their preodp omsethodology was based on linear regression of variety yield on experimental meaenld y in order to observe varietal stability across varying environments (Finlay and Wilkinso1n9,6 3; Eberhart and Russell 1966). Stability analysis has been adapted for use in caorimngp agronomic treatments across different environments consisting of the linear regressio tnre oaftment mean yield on the environmental mean (Raune t al., 1993). The concept of stability has been def iine md any ways by many researchers and several biometrical methods, iinncglu udnivariate and multivariate (Lient al., 1986; Becker and Leon, 1988; Crossa, 1990). Sttya binildi ices are usually univariate while a genotype’s response to different environments inss icdoered to be multivariate (Lient al., 1986). Through multivariate analysis, genotypesh wsimt ilar responses are observed to cluster together (Crossa, 1990). There are two types olof tbsi pthat have been widely used to visualize GEI and these are the AMMI and, genotype and gepneo xty environment interaction (GGE) biplots (Gauch, 1988; Gauch and Zobel, 1997; Yeta nal ., 2000; Mae t al., 2004). The main difference between the two approaches is that GiGplEot banalysis is based on location centred PCA while AMMI analysis is observed as double cedn trPCAs. When a number of environments and genotypes are involved it is nlwoat yas easy to visualize ‘which won where’ in the AMMI biplot and at times it could be decevpet,i as suggested by Yaent al. (2007). Still AMMI is regarded as a better tool for presentinngc cluosions rather than a tool for determining “which won where”. To ensure that suitable varieties are recommenodre cdu fltivation by farmers in the country, the Agricultural Technology Clearing Committee (ACT)T developed guidelines for release of varieties in Malawi. The most important rule ina rteiol n to GxE is that the candidate variety must have been evaluated for three years and shstoawbnle performance at both on-farm and on-station trials. If it is an introduction, it mtu hsave two years of data for on-farm and on- 121 station trials in the country, supported by regilo dnaata. In addition to this rule, the variety release dossiers should clearly specify the enmvireonnt it is suited for e.g. low-altitude or mid- altitude. In this case, multivariate stability aynsaisl like AMMI and GGE are important tools in selecting a stable variety and making recommteionndsa as to where it is suitable for cultivation. Stable performance of a variety is also dependne ncte ortain genetic properties. Zivanoveitc al. (2004) indicated that breeding o1f Fhybrids of maize is successful because it exp loits heterosis and increases homogeneity. The unifo rmofi tyhybrids consists of: (i) genetic homogeneity and (ii) genetic stability. Genetic hoogmeneity is focused on maintenance of the identity of genotypes, while genetic stability tesn tdo maintain homeostasis (phenotypic uniformity) in different environments. The level yoifeld depends on genetic yield potential (all favourable genes incorporated into a cultidvuari ng a breeding process). Stability of yield or of any other trait depends on the ability of ivae ng cultivar to react to changes in the environment which it is subjected to; this is tedrm aes phenotypic plasticity (Frey, 1983). In the present study, the objective was to study txhEe Gand stability of the tropical and sub- tropical maize genotypes in low pH, low N and opatli mconditions. 5.3 Materials and methods The experimental design, experimental materials saitned description are given in Chapter 4 Section 4.3. The low N trial of 2012 was destroybeyd l ivestock just before harvest, hence cluster analysis alone combining with low N dat a2 0o1f 3 was possible using environmental means. The reason for evaluating low N tolerances twoa check if other genotypes had additional attributes for this stress apart fromat twhhich were pre-described as originating from a low N tolerant source e.g. LOW N POOL C3A-B .g enotype which has additional attributes is better placed for high adoption brym fears e.g. a released variety ZM523 is drought tolerant with low N tolerance as an additionali bautttre and is recognised through the quantities of annual certified seed sales, as widely cultidv actoempared to other OPVs available in the country. 122 5.4 Data analysis 5.4.1 Analysis of variance Statistical analyses were performed using varioouftsw sare packages: GenStat th1 V6ersion (2013), Agrobase (2010) and NCSS (Hintze, 2007)e. TAhMMI model, which combines ANOVA with PCA, was used to study the nature of G. GExIE was partitioned into sources of variation (i) additive main effects for genotypensd aenvironment and (ii) non-additive main effect due to GEI. 5.4.2 Stability analysis AMMI analysis for mean yield was performed usingr oAbgase (2010). GGE biplot analysis was conducted using the GGE biplot in GenStat ()2.0 T1h3e model for GGE biplot (Yan and Hunt, 2002) based on single value decompositionD ()S oVf the first two principal components (PC) was used. 5.5 Results 5.5.1 Analysis of variance for additive main esff mecutltiplicative interaction The combined ANOVA of the 45 maize genotypes evtaeldu afor two years across eight locations according to the AMMI model is presenitne dT able 5.1. The ANOVA indicated highly significant effects (p<0.01) for environmse,n tgenotypes and GEI. The IPCA’s were ordered according to decreasing importance. Thees tF w-tas highly significant (p<0.01) for the first six IPCA axes and at p<0.05 for the sevenPtCh AI . The total variation explained ranged from 2.5% for genotypes, 59.4% for environments 1a7n.d5% for GxE. The variation due to GxE was over five times the variation due to gepneosty as main effects. The first six IPCA axes explained 75.3% of the GEI. The first IPCAt ucarepd 32.8% of the total interaction sum of squares in 8.8% of the interaction degreese oefd for m (GxE). 123 Table 5.1 AMMI Analysis of variance for grain yi efoldr two years 2011/12 and 2012/13 Total variation % GxE Source s DF SS MS explained (% ) Eigenvalue s Explained Cumulative % Total 2159 503381378 2 Environment s 15 298964743 7 199309829.1** 59.4 Reps within En v 32 159985978. 2 4999561.8 Genotype 44 124590289. 1 2831597.5** 2.5 Genotype x En v 660 879765257. 9 13329877.7** 17.5 IPCA 1 58 288728372. 6 4978075.4** 96242790.9 32.82 32.82 IPCA 2 56 136931066 2445197.6** 456436688. 7 15.56 48.38 IPCA 3 54 88841154.8 1645206.6** 29613718.3 10.10 58.48 IPCA 4 52 75415838.9 1450304.6** 25138613.0 8.57 67.05 IPCA 5 50 72193861.2 1443877.2** 24064620.4 8.21 75.26 IPCA 6 48 54027784.3 1125578.8** 18009261.4 6.14 81.40 IPCA 7 46 47403253.8 1030505.5* 15801084.6 5.39 86.79 Residua l 1408 879824820. 1 624875.58 Grand mean- 2516.39, R- Squared 0.8252, CV = 31. 4*%P<, 0.05, **P<0.01; IPCA= Interaction principal component axis =C cVoefficient of variatio n, 124 The second IPCA explained 15.6% of the interactsiounm of squares in 8.5% of the interaction degrees of freedom (GxE) (Table 5.1). Results for IPCA1 and IPCA2 are presented in Ta5b. l2e and Table 5.3 and are sorted in terms of environmental mean. G20 ranked first olvl.e Irna the biplot in Figure 5.1, low pH environments: Bembeke Trurn Off (BKET), Bembekei cOef,f (BKEO), Lunyangwa (LUN) and Tsangano (TSA) were distributed with the lowyierld ing environments in quadrants 1 (top left) and IV (bottom left) (Figure 5.1), wleh ioptimal environments: Chitala (CLA), Chitedze (CZE), Meru (MRU) and Baka (BKA) were ptioosnied with the high yielding environments in quadrant II (top right) and III t(tboom right). The genotypes categorised under favourable environments with above averagaen ms ewere G20, G18, G13, G26 and G41. Among them G20 was found to be most stable2. wGa4s identified as the most unstable genotype (quadrant IV) and the other gyepneost under low yielding environments are shown in the lower left quadrant of the bip Wlotit.h respect to the environments, closer relationships were observed among Bembeke Tur(nB oKfEf T), Bembeke Office (BKEO), Lunyangwa (LUN) and Tsangano (TSA). Chitedze (CZwEa)s identified as a suitable environment as its IPCA score and vector was noe tahre t origin (zero). 5.5.2 Genotype and GEI scatter biplot and polygioewn vof grain yield across eight environments for 20011/12 and 2012/13 The polygon was constructed from genotypes G20,, G452, G2, G40, G16 and G8 as markers (Figure 5.2). Eight lines were drawn sntagr ftriom the origin and extended beyond the polygon such that the biplot was divided initgoh et sectors and environments fell into three of them. Bembeke Turn off (BKET) fell in soerc 1t delineated by rays 1 and 2 and the vertex genotype was G16. Similarly one envireontm, Bembeke Office (BKEO) fell into sector three and the vertex genotypes were aGn2d8 G8. All optimal sites fell into sector four and were delineated by rays four avned afind the vertex genotype was G20. The remaining two low pH sites: Tsangano (TSA) aLnudn yangwa (LUN) fell just at the origin. 125 Table 5.2 IPCA1 and IPCA2 scores for the top 20o gtyepnes based on mean grain yield at eight locations for two seasons Mean yield No Entry code kg ha-1 IPCA1 IPCA2 1 G20 2977 23.25322 3.15773 2 G27 2833 7.16857 -6.46885 3 G30 2729 18.05689 -16.53917 4 G8 2728 18.77411 -10.69649 5 G5 2719 15.12165 -8.60019 6 G22 2663 -4.61057 2.08058 7 G33 2646 19.09816 -2.1516 8 G28 2620 0.96947 -4.78308 9 G39 2614 5.95664 2.00067 10 G32 2595 5.91891 13.04379 11 G36 2594 2.29902 17.73102 12 G26 2593 3.39991 1.41781 13 G45 2591 5.21745 9.87007 14 G18 2586 0.03059 3.77821 15 G43 2584 3.63716 14.94491 16 G6 2580 -7.02094 -12.23653 17 G10 2565 10.19885 2.24328 18 G12 2552 5.78089 13.91971 19 G15 2532 -1.59902 -5.52944 20 G44 2522 11.0564 12.81569 Table 5.3 IPCA1 and IPCA2 scores for the eight reonnvmi ents, ranked based on environmental mean for two seasons Environment Environmental mean IPCAe[1] IPCAe[2] MRU 3964 22.77 14.34 CLA 3618 47.07 -9.58 BKA 3350 10.59 44.64 CZE 2802 -31.12 2.85 BKET 2461 14.44 -38.27 BKEO 1927 -20.15 -8.73 TSA 718 -21.26 -0.35 LUN 516 -22.35 -4.89 MRU = Meru, CLA = Chitala, BKA = Baka, CZE = Chitedz eB,keT = Bembeke Turn Off, BkeO = Bembeke Office, TSA = Tsangano Low pH site, LUN = Lunyangwa 126 Adapted /high yielding Low yielding High yielding Non- adapted Figure 5.1 AMMI biplot for yield for genotypes aenndv ironments across two seasons 2011/12 and 2012/13 MRU = Meru, CLA = Chitala, BKA = Baka, CZE = Chitedz eB,keT = Bembeke Turn Off, BkeO = Bembeke Office, TSA = Tsangano Low pH site, LUN = Lunyangwa 127 Unstable Stable Figure 5.2 Genotype and GEI scatter biplot and gponly view of grain yield across eight environments for 20011/12 and 2012/13 se asons MRU = Meru, CLA = Chitala, BKA = Baka, CZE = Chitedz eB,keT = Bembeke Turn Off, BkeO = Bembeke Office, TSA = Tsangano Low pH site, LUN = Lunyangwa 128 5.5.3 GGE comparison biplot across optimal andp lHo wen vironments combined for two seasons The GGE biplot for ranking of environments based doisncriminating ability and the representativeness across environments relatiyvie ltdo p erformance is presented in Figure 5.3. The ideal environment was positioned nearc ethnetr e of the average environment axis (AEA) which is represented by a small circle nehaer et nd of the arrow. The arrow at the end of AEA points towards a direction indicatinge tmhost informative location. The biplot indicated that environment Chitedze (CZE) was thoes tm representative and discriminative in terms of grain yield performance based on AEnA t.e Irms of genotypes G38, G18 and G26 were close to the small circle (ideal environtm). e 5.5.4 Ranking of genotypes based on both mean a ynidel dstability view of the GGE biplot GGE biplot of the genotypes based on both the maneda ns tability showed the relative mean performance and stability of hybrids across sea s(Foingsure 5.4). The genotypes G20 and G33 were high yielding based on average environ mcoeonrtdination abscissa (AECa) and average environment coordination ordinate (AECoh)e. Tgenotypes G18, G13, G26 were near to the AEA which shows that they are very lest abbut not the highest yielding genotypes. 5.5.5 Cluster analysis of maize genotypes ando ennmviernts Clustering of genotypes at cut off point of 1.0 dpurcoed four clusters (Figure 5.5). Cluster I consisted of 10 genotypes (G1, G14 up to G9)c alunsdt er II consisted of seven genotypes (G11, G2 up to G7), cluster III consisted of 27 ogteynpes (G10, G12 to G8) and cluster IV consisted of one genotype G20 which was ranked neur monbe by AMMI in terms of mean grain yield performance. Cluster analysis of ennvmiroents (Figure 5.6) at cut off point 1.0 with a cophenetic correlationc o(pr = 0.84) produced three clusters. Cluster I coends isotf three stress sites, the low N site, Lunyangwa asnadn gTano. Cluster II consisted of two low pH stress sites Bembeke Office and Bembeke Turn, COluffster III consisted of four optimal sites Baka, Chitedze, Chitala and Meru. 129 Average Environ ment Axis (AEA) Figure 5.3 Genotype and GEI comparison biplot oaifn g yrield across eight environments for 2011/12 and 2012/13 MRU = Meru, CLA = Chitala, BKA = Baka, CZE = Chitedz eB,keT = Bembeke Turn Off, BkeO = Bembeke Office, TSA = Tsangano Low pH site, LUN = Lunyangwa 130 Average Environ ment Axis (AEA) Figure 5.4 Ranking of genotypes based on both myieladn and stability view of the GGE biplot MRU = Meru, CLA = Chitala, BKA = Baka, CZE = Chitedz eB,keT = Bembeke Turn Off, BkeO = Bembeke Office, TSA = Tsangano Low pH site, LUN = Lunyangwa. 131 Figure 5.5 Dendrogram of 45 maize genotypes as revealed byM UAP cGluster analysis based on AMMI adjusted mean yields comdb fionre two seasons using Euclidean distance and standard deviation as sgc maleinthod 132 Environment Low N Tsangano Lunyangwa Bembeke T. Off Bembeke office Chitala Chitedze Meru Baka 2.00 1.67 1.33 1.00 0.67 0.33 0.00 Dissimilarity Figure 5.6 Dendrogram of nine environments as rledv ebay UPGMA cluster analysis based on environmental means and Euc ldidisetaannce and standard deviation as scaling method 133 5.6 Discussion Environments and genotypes were both plotted atso rvse cand points respectively on the AMMI biplot. Genotypes and environments that wenr ec liose proximity are considered to be similar in terms of performance and discrimionna toi f genotypes. The angle between two vectors indicated the degree of associationc oorre lation. Small angles indicated similarity, a 90o angle indicated orthogonality and no associatinodn an angle >9o0 indicated a negative correlation. The sites Lunywaan gand Tsangano were close to each other and these had the lowest pH. Genotypes GH8 p(oLpP9), G28 (LPHpop10) and G30 (LPHpo3) were in close proximity and these were lpoHw tolerant populations from CIMMYT-Colombia and this demonstrated similarity tohfe genotypes. The orthogonal projections of genotypes on environment vectorsic aintde the relative performance of genotypes in a given environment: that is, the tgere tahe projection of the genotype in the positive direction, the better the performanceh oaft tgenotype in that environment. Drought tolerant (DT) genotypes were close to each othe4r4 [(GZM523), G12 (TZE YDT STR C4- B), G43 (ZM309) and ZM721]. These genotypes arel yli kto have some genes in common that were at play. Tsangano, Lunyangwa and BemObeffkicee sites were close to each other and these were classified in one group as acidteics ,s bi y use of the Soil Interpretation Guide (Hornecke t al., 2011). These sites are also associated with peorofor rmance in terms of maize yields such that 85% yield reduc twioans reported for Lunyangwa (Munthali and Chilimba, 2004). When the environment contributes a large percen otaf gveariation, it implies that it was a major factor that influenced yield performance a(I,s 2s009). When plotting the genotypes and the environments on the same graph, the astisoonc biaetween the hybrids and the environments can be seen clearly. IPCA scoresg oefn ao type in the AMMI analysis are an indication of the stability of a genotype over ernovniments (Gauch and Zobel, 1996; Purchase, 1997). The greater the IPCA scores, r epitohseitive or negative, the more specifically adapted a genotype is to certain eonnvmirents sampled. For instance G27 (Low N Pool C3-B) was close to CZE (Chitedze), indicga ttihnat it is a low N tolerant genotype and is able to do much better under ideal condsit ipornobably because of its high N use efficiency (NUE). The present study recorded gepneo tcyontribution to total sum of squares as very low at 2.5% and was consistent with finsd ibngy Babice t al. (2010) who reported 134 2.2% but was not consistent with findings by Mitirćo vet al. (2011) who reported a somewhat higher 9.17% contribution. The genotypes categorised under favourable envieronntsm with above average means were G20, G18, G13, and G41. Among them G20 (LPHpop2a1s) fwound to be more stable and this was from the low pH tolerant populations froCmIM MYT-Colombia. Its tolerance to low pH might have contributed to its stable perfaonrmce across seasons and all locations tested. Genotypes grouped under low yielding envmireonts are shown at the lower left quadrant (IV) of the biplot. G42 (VPO97) from CIMMTY-Zimbabwe was the most unstable genotype identified by the AMMI model (uFrieg 5.1). Genotypes that are close to each other tend to have similar performance ands et htohat are close to a specific environment indicate their better adaptation tot pthaarticular environment. The five IPCA axes can be taken as adequate dimensions of tah,e h doawtever, only the first two IPCA axes were plotted on the biplots to help inveseti gthaet GEI pattern for each genotype. The biplot showed that Chitala was the most discrimininga et nvironment for the genotypes as indicated by the longest distance between its mr aarnkde the origin and gave information of the performance of the genotypes. This site awsasso ciated with genotype G8 (LPHpop 9) which might be specifically adapted to this .s Citehitala was used as an optimal site in terms of pH but it is used as random drought scinrege snite classified as low-land tropical zone E by CIMMYT (2008). G8 (LPHpop9) coincidenyta wllas among the low pH tolerant populations from CIMMYT-Colombia and it might bele torant to random drought. Piñeros et al. (2005) and Hartwige t al. (2007) reported that low pH negatively influencthees use of soil nutrients and induces plants to be morec esputsible to drought. It implies that a genotype which is tolerant to low pH is likely toe bless susceptible to random drought. However, due to its high IPCA scores, genotypea vbailritiy at this environment may not exactly reflect the average genotype performancroes as cenvironments. This is consistent to the findings by Arulselvi and Selvi (2010). Ftohre environments, a closer relationship was observed between Bembeke Turn Off, Bembekec eO, fLfiunyangwa and Tsangano low pH site. Chitedze was identified as a stablvei reonment as its IPCA score and vector is near to the origin (zero). This site is in thied -maltitude ecology and is in the arable land of Lilongwe Plain with a longer rainfall season nth Ca hitala and these results have confirmed that the country’s best yields are obetda inin the mid-altitude ecology (MOA, 1994). Generally genotypes with a smaller vectogrl ea nhave a similar projection, which 135 designate their proximity in grain yield and pemrfoarnce. Those genotypes that are clustered close to the centre tend to be stabl eth aonsde far apart unstable. The GGE biplot construction is carried out by pinlogt tthe first PC scores of the genotypes and the environments against their respective s cfoorre PC2 that result from SVD of environment-centered or environment-standardisendo tgyepe-by-environment data (GED) (Yan et al., 2000; Setimelae t al., 2007). According to Setimeleat al. (2007), the purpose of the polygon view of the biplot is basically toh osw which hybrids won in which environments. The genotypes located furthest frhoem b tiplot origin demarcate the corners of the polygon. The perpendicular lines that araew dnr from the biplot origin divide the biplot into sectors or mega environments. The eonvmirents are contained into different sectors such that different sectors contain dinffte wreinner genotypes (Yaent al., 2007). The results showed that the polygon was constru fcrotemd genotype markers G20, G45, G42, G2, G40, G16 and G8. Among these markers, iGn 4th0e upper left quadrant (I) marked the unstable environment while G42 in thwee lro bottom left (IV) marked the low yielding environment. Depending on the objective t hoef research, selection could be carried out in the quadrant with markers G20 (set aebnlvironment) and G45 (high yielding environment). Eight lines were drawn starting frothme origin and extended beyond the polygon such that the biplot was divided into e isgehcttors and environments fell into three of them. Bembeke Turn off fell in sector 1 delineeda tby rays 1 and 2 and the vertex genotypes were G16. Similarly environment Bembefkfeic eO, fell into sector three where the vertex genotypes were G28 and G8. All optimitaels sfell into sector four and were delineated by rays four and five and the vertexo gtyepnes was G20. The remaining two low pH sites: Tsangano and Lunyangwa fell just at trhigei no. The construction of GGE comparison biplots is dwoniteh the aim of ranking environments based on discriminating ability and the represeivnetnaet ss across environments relative to yield performance (Setimeleat al., 2007). In this type of biplot, the ideal enviroennmt is one positioned near the centre of the AEA whicrhe pisr esented by a small circle near the end of the arrow as indicated in Figure 5.3. Thney tciircle in the biplot represents the average environment and is defined by the averaCg1e aPnd PC2 scores across the environments. In the present study, the biplotc iantdeid that environment Chitedze was the 136 most representative and discriminative in termsg roafi n yield performance based on the AEA and genotypes G38, G18 and G26 were close et os mthall circle showed that they were stable but not the highest yielding genoty pes. The AEA is the average environment axis and theje pctrion of the genotypes onto this line represents the main effects of the genotySpetsi m( elae t al., 2007) therefore the AEAa ranks the genotypes according to the mean perfocrem wanhere by a small a is for abscissa . Usually ranking of the genotypes on the AEA iso acsiated with the genotype main effect that is, the AECa approximates the contributione aocf h genotype to the main effects of genotypes. Similarly, AECo expresses the genotyopnet ricbutions to GxE and thus it represents the genotype stability across environtsm wehnere by a small o is for ordination. The AEA from the biplots points towards maize geynpoets with high and stable mean grain yield across environments. The genotypes G20 an3d a Gre3 high yielding based on AECa and AECo. An “ideal” genotype may not exist in ptircaec but it can be used as a reference for genotype evaluation (Mitroćv iet al., 2011). A genotype is more desirable if it is loecda t closer to the “ideal” genotype (Kayeat al., 2006). Plant breeders prefer genotypes that are high yielding and also stable across environmeInt st.h is ranking biplot, G42 was identified as the most unstable genotype as it lowcaast ed far from the “ideal genotype”. The GGE biplot explained 68.6% of the G+GE varina.t iAoccording to Yane t al. (2007), the greater the contribution, the more confidenhce rtesearcher would have in the interpretations based on the biplot. However, sifm aa ller portion of the total variation is explained, it does not necessarily mean that tphleo tb iis useless. GGE biplots provide more information than AMMI biplots while the latter gisv eimportant details in the ANOVA, making both useful and both should be used in lsittya bainalyses. According to Ghaderrei t al. (1980) cluster analysis is the most widely usecdh nteique for classifying environments or genotypes into homogoeuns egroups. It operates on a matrix of dissimilarity or dissimilar indexes for all poibslse pairs of genotypes or pairs of environments, on which it is being clustered. Ine pthresent study, cluster analysis was performed to study the pattern of groupings andir oenmv ents. The hierarchical clustering report indicated that cluster I consisted of 10o gtyepnes (G1, G14 up to G9) and cluster II consisted of seven (G11, G2 up to G7), clustecro InIIs isted of 27 genotypes (G10, G12 up 137 to G8) and cluster IV consisted of one genotype0, Gw2hich was ranked number one by AMMI in terms of mean grain yield performance. Tehfeorre both methods could be used to identify stable genotypes. Cluster analysis of environments at cut off poi.n0t w1ith a cophenetic correlation ocfo pr = 0.84 produced three clusters (Figure 5.6). Thea hrciehrical clustering report indicated that cluster I consisted of three environments, low tNe, sLiunyangwa and Tsangano. Cluster II consisted of two low pH stress sites, Bembeke Oe fafincd Bembeke Turn Off. Cluster III consisted of four optimal sites Baka, Chitedze, taClah i and Meru environments. This technique could be useful in general grouping ocaf tlioons that have stresses and further analysis is required to identify specific stresaste psl ay. 5.7 Conclusions The low pH soil environment had an effect on gryaieinld performance such that the low pH sites were positioned in the low yielding arne ath ie AMMI biplot. Chitala optimal site was identified as the most discriminating environmt feor the genotypes, and the most stable environment. Genotypes G20 (LPHpop21), GV1P8O (52), G13 (VPO721) and G41 (VPO96) were positioned in the favourable environtms ewith above average means. Cluster analysis results were similar to AMMI rets usluch that cluster III comprised of genotypes that were stable, except for G2O (LPH1p)o tph2e most stable, which was in its own cluster. G42 (VPO97) was identified as the m uonsttable genotype. GGE biplots explained more variation and gave minofroer mation than AMMI while the latter gave important details on the ANOVA, maktihneg m both useful and both should be used in stability analyses. Similarly clustering e onfvironments classified the lowest pH sites in cluster I and intermediate pH sites ins tcelur II. All optimal sites were positioned in the third cluster. Clustering identified the l oNw site as similar to the lowest pH sites in terms of its environmental mean. 138 5.8 References AGROBASE 2010. Agronomix Software, Inc. GeneratIioI nu ser’s Manual. Version II. Revised Edition. www.agronomix. Agronomix softwa Wrei,nnipeg, M.B, Canada. Aina, O.O., A.G.O. Dixon and E.A. Akinrinde. 2007A.d ditive main effects and multiplicative interaction (AMMI) analysis for yide lof cassava in Nigeria. Journal of Biological Sciences 7: 796-800. Annicchiarico, P. and M. Perenzin. 1994. Adapta tpioantterns and definition of macro environment for selection and recommendation of mcomn wheat in Italy. Plant Breeding 113: 197-205. Arulselvi S. and B. Selvi, 2010. Grain yield staitby ilof single cross maizeZ (ea may sL.) hybrids over three different environments. Elecitcr oJnournal of Plant Breeding 1: 577-584. Babic, V., M. Babic, M. Ivanovic, M. Kraljevic-Balliac and M. Dimitrijevic., 2010. Understanding and utilization of Genotype by Ennvmiroent Interaction in Maize Breeding. Genetika 42: 79-90. Becker, H.C. and J. Leon. 1988. Stability analyins ips lant breeding. Plant Breeding 101: 1-23. Brewbaker, J.L. 1985. The Tropical environment mfoari ze cultivation. In: A. Brandolini and F. Salamini (Eds.) Breeding Strategies for Me aPirzoduction Improvement in the Tropics. Food and Agriculture Organisations andti tIuntso Agronomico I’Oltremarc, Firenze, Italy, pp. 47-77. CIMMYT, 2008. Characterisation of Maize Germplasmro wGn in Eastern and Southern Africa. Results of the 2008 Regional Trials Cooardteind by CIMMYT. Harare. Crossa, J. 1990. Statistical analysis of multiliocna trials. Advances in Agronomy 44: 55- 85. Eberhart, S.A. and W.A. Russel, 1966. Stabilitya pmaerters for comparing varieties. Crop Science 6: 36-40. Finlay, K.W. and G.N. Wikinson, 1963. The analyosifs adaptation in plant breeding programme. Australian Journal of Agricultural Rersceha 14: 742-754. Frey, K.J. 1983. Plant population Management anede dbirng. In: D.R. Wood (Ed.) Crop Breeding, ASA, Madison, USA, pp.55-88. 139 Gauch, H.G. 1988. Model selection and validationr yfoield trials with interaction. Biometrics 44: 705-715. Gauch, H.G. and R.W. Zobel. 1996. AMMI analysisy ioefl d trials. In: M.S. Kang and H.G. Gauch (Eds.) Genotype by environment interaction8 5p-1p22.CRC Press, Boca Raton, New York. Gauch, H.G. and R.W. Zobel. 1997. Identifying meegnav–ironment and targeting genotypes. Crop Science 37: 311-326. GenStat. 2013. Introduction to GenStat for Windo1w6s.1 th Edition. VSN International, Hemel Hempstead, Hertfordshire HPI. IES, UK. Ghaderri, A., E.H. Everson and C.E. Cress. 198a0s. sCiflication of environments and genotypes in wheat. Crop Scien 2c0e: 707-710. Hallauer, A. R. and J. B. Miranda. 1988. Quantvitea tgienetics in maize breeding. Iowa State Univ. Press. Ames, IA. Hartwig, I., A.C. Oliveira, F.I.F. Carvalho, I. Btearn, J.A.G. Silva, D.A.M. Schnidt, I.P. Valerio, L.C. Maia, D.A.R. Fonseca and C.E.S. R 2e0is0.7. Mecanismos associados à tolerância ao alumínio em plantas. Semina: Caiêsn Acgi rárias 28: 219-228. Hintze, J.L., 200.7 Number Cruncher Statistical System (NCSS). Stcaatils tSi ystem for Windows. Kaysville, Utah. Horneck, D.A., D.M. Sulivan, J.S. Owen and J.M. tH., a2r011. Soil Test Interpretation Guide. Oregon State University Extension Service. Issa, A.H. 2009. Genotype by environment interanc atinod yield stability of maize of maize hybrids evaluated in Ethiopia. MSc thesis subm itote tdhe University of Free State. South Africa. Kang, M.S., D.P. Gorman and H.N. Pham. 1991. Apaptiloicn of a stability statistics to international maize yield trials. Theoretical anpdp Alied Genetics 81: 162-165. Kaya, Y., M. Akcura and S. Taner. 2006. GGE bipalnoat lysis of multi-environment yield trials in bread wheat. Turkish Journal of Agricurelt uand Forestry 30: 325-337. Lin, C.S., M.R. Binns, and L.P. Lefkovitch, 1986ta. bSility analysis: Where do we stand? Crop Science 26: 894-900. Ma, B.L., W. Yan, L.M. Dwyer, J. Fregeau-Reid, H .VDo. ldeng, Y. Dion and H. Nass. 2004. Graphic analysis of genotype, environmenttr,o gnei n fertilizer and their interaction on spring wheat yield. Agronomy Jou r9n6a:l 169-180. 140 Malawi Government. 1994. Guide to Agricultural purocdtion in Malawi 1994. Extension Aids Branch. Ministry of Agriculture. Lilongwe. Mitrović, B., D.S. Tanisavljevi, S. Treski, M. Stojakćo,v iM. Ivanović, G. Bekavac, M. Rajković, 2011. Evaluation of Experimental Maize Hybridss tTeed in Multi-location Trials using AMMI and GGE Biplot Analyses. TurkisJho urnal of Field Crops 17: 35-40. Munthali, M.W. and A.D.C. Chilimba. 2004. Effectfs c oomposting materials and methods of composting on quality of compost and maize ysi.e lIdn: 2004 Annual Report of Soils and Agricultural Engineering Commodity Gro uMpi,nistry of Agriculture, Chitedze Research Station, Lilongwe, Malawi. pp. . 52 Myers, G.O. 2004. Estimation of potential breedivnagl ue and genotype stability. http//www.cottoninc.com. Piñeros, M.A., J.E. Shaff, H.S. Manslank, V.M. Caalhrvo Alves and L.V. Kochian. 2005. Aluminum Resistance in Maize Cannot Be Solely Eixnpelda by Root Organic Acid Exudation. A Comparative Physiological Study. P lPahnyt siology 137: 231-241. Purchase, J.L., 1997. Parametric analysis to dbees cGr ix E Interaction and yield stability in winter wheat. PhD Thesis. Department of agron,o mFayculty of Agriculture, University of Free State, Bloemfontein, South Aafr.i c Raun, W.R., H.J. Barreto and R.L. Westerman, 19U9s3e. of stability for long term soil fertility experiments. Agronomy Journal 85: 159-1. 67 Ramagosa, I. and P.N. Fox, 1993. Genotype x Enmvieronntal interaction and adaptation. In: M.D. Hayward, N.O. Bosemark and I. Ramagosa s.()E dPlant Breeding: Principles and prospects. Chapman and Hall, Lon pdpo.n 3, 73-390. Setimela, P.S., B. Vivek, M. Bänziger, J. Crossda Fa.n Maideni. 2007. Evaluation of early to medium maturing open pollinated maize varieitnie sS ADC region using GGE biplot based on SREG model. Field Crops Resear3ch: 11601-169. Yan, W. and L.A. Hunt. 2002. Biplot analysis of ldleial data. Crop Science 42: 21-3 0. Yan, W. and N.A. Tinker. 2006. Biplot analysis ouf lmti-environment trial data: Principles and applications. Canadian Journal aonf tP Sl cience 86: 623-645. Yan, W., L.A. Hunt, Q. Sheng, and Z. Szlavnics. 02.0 C0ultivar evaluation and mega- environment investigation based on GGE biplot. C Srocpience 40: 596-605. Yan, W., M.S. Kang, B. Ma, S. Woods and P.L. Conliurnse. 2007. GGE Biplots vs AMMI analysis of genotype by environment data. Cropn Sccei e41: 656-663. 141 Yates F. and Cochran W. 1938. The analysis of gsr ouf pexperiments. Journal of Agricultural Science 28: 556-580. Zivanovic, T., I. Maja Vracarevic, S. Krstanovicd a Gn. Surlan-Momirovic. 2004 Selection on Uniformity and Yield Stability in Maei.z Journal of Agricultural Sciences 49: 117-130. 142 CHAPTER 6 Evaluation of diallel crosses for combining a bielittwyeen selected tropical and sub-tropical maize lines for low pH tolerance 6.1 Abstract This study was conducted to assess the GCA efofef cptsa rental maize inbred lines and SCA effects for the diallel crosses for yield aniedl dy related traits and explore their use in hybrid development for low pH tolerance. Sixty sFi1xs generated by crossing 12 inbred lines were evaluated alongside two hybrid check sfo uart locations in Malawi. The experimental design was a (0.1) alpha lattice wthitrhe e replications. Positive and highly significant GCA effects for grain yield were obsedrv for inbred line CZL999601 across low pH and optimal conditions. Negatively and sfiigcnaint GCA effects for grain yield were observed for inbred line CML161 across lowa pnHd optimal conditions. Single cross hybrids CZL999601/CML144, CML144/CML202, CML481/CM28L8 and CML161/CM172 were identified as good specific conmebrsi for grain yield. The estimates of broad sense heritability were high for days 0to% 5 anthesis, anthesis-silking interval, grain yield and plant vigour. At a cut-off point 1o.f0 the UPGMA clustered the inbred lines based on GCA for grain yield into two mainu sctel rs through use of Euclidean distance and standard deviation as type of scahler.e T was a high cophenetic correlation of rcop = 0.87 and the pattern mostly followed the originth oef maize inbred lines such that six out of seven inbred lines from CIMMYT-Colomb(itar opical) were grouped in the second cluster. The inbred lines and specific conmatbioins identified in this study will be used in the National Maize Breeding Programme efovre dlopment of genotypes tolerant to low pH and diseases for yield improvement and squubesnet food security in the count ry. 6.2 Introduction Maize is the world’s most widely grown cereal ansd t hie primary staple food in many developing countries (Morriest al., 1999). In Malawi, two main types of maize cualtrisv are cultivated by farmers, hybrids and OPVs. TheV sO Pinclude synthetics, composites, 143 and local types which are locally called Chamak Ionlo t.hese types of maize, hybrids have the highest yielding potential, followed by syntihcse tand composites and then lastly local varieties. The differences are based on the brge epdrioncess in case of hybrids, synthetics and composites. The performance of maize genotuyspeds as parents is evaluated based on the performance of the cross progeny. With rcets tpoe hybrids, the parental lines are chosen based on their SCA as measured in theirid h oybffrspring. Usually two or three selected inbred lines are used in developing s-icnrgoless and three-way cross hybrids, respectively. Synthetics are developed by inter-crossing a la nrguemrber of selected parental lines of known superior combining ability, for example indb relines which are known to give superior hybrid performance when crossed in all bcionmations. A composite is developed by selecting parental lines of relatively similaar tmurity periods and mixing the seed from the individuals. In both cases, seed is maintabinye adl lowing full-sib pollination (or plant to plant pollination in the same population). Th etyspees are also referred to as open- pollinated because there is no control of pollenv emmoent when planted in isolation for seed increase (bulking), especially for basic aenrdti ficed seed multiplication. In other words, some form of inbreeding occurs as thereo ids en-tasselling of the male flowers of the female parent as in hybrid crosses. Hence frasr imn eMalawi are advised to recycle the OPVs for two seasons and the third season theyl ds hloouk for new seed because of inbreeding depression and genetic drift which osc cwuirth time of recycling the seed and reception of foreign genes from other varieties. Crop estimates for 2012 showed that hybrids countteridb 56.5% to the national total production while composites and synthetics conteridb u30.5% and local varieties the least at 13%. In terms of hectarage, hybrids contribu4t0e%d , composites/synthetics 30% and local varieties 30% (MOA, 2012). Combining abilisty c onsidered in hybrids more than in synthetics which are based only on GCA. The mosllt- kwneown synthetic and composite varieties released by the National Maize Breedirnogg rPamme in Malawi are Masika and Chitedze Composite A, respectively. In terms of rhidysb, it is Malawi maize hybrids number 17 and 18 (MH17, MH18) and these are thset ffliirnt three-way cross hybrids which are preferred because of their good pounldit ya. bTihe two hybrids had a common male parent which is very flint and was identifiaesd a good combiner for flintne ss. 144 The concept of GCA and SCA was introduced by Sper aagnud Tatum (1942) and its mathematical modelling was done by Griffing (195 T6h)e. value of any population depends on its potentialp er se and it’s combining ability in crosses (Vacaerot al., 2002). The variances of GCA and SCA are related to the typ gee onfe action involved. Variance for GCA includes the additive portion while that of S CinAcludes the non-additive portion of total variance arising largely from dominance apnids teatic deviations (Rojas and Sprague, 1952). GCA and SCA are powerful tools used by p blarenet ders in selecting the best parents for further crosses. Studies on combining abilietylp h breeders in identifying parental lines with good GCA and in detecting hybrids with goodA S (CNdhela, 2012). Devi and Singh (2011) indicated that the best performing genot yopuegsht to show stable performance across environments in multi-environment trials.w Heover, it is important to bear in mind that heritability is not only influenced by thei t ruander consideration but is also influenced by the population and environmental conditions wh hinicdividuals are exposed to, as well as the method of data collection used (Falcone rM aancdkay, 1996). In terms of types of genotypes, reports indicaatet rtehsearchers in the late 1950s and early 1960s recommended the use of maize genotypes wbirtoha ader genetic base, because they are more tolerant to stresses than the narrow igce bnaest e genotypes. Lerner (1954), Lewontin (1957) and Allard and Bradshaw (1964) retepdo that yield stability is influenced in part by the genetic structure of the varietyh s uthcat more heterogeneous varieties are less affected by environmental differences. Spr aagnude Federer (1951) found that maize double-cross (DC) hybrids have smaller GEIs and maorere stable than single-cross hybrids. Valdivia-Bernal and Hallauer (1991) alseop orrted that more homogeneous generations, inbred lines and1’ sF had larger GEIs than more heterogeneous geonnesra, ti F2 and backcross populations. DC hybrids are rarseelyd unowadays because of the cost of producing the seed, but with the current maize upcrotiodn constraints, DC hybrids and OPV populations are better placed and the proce sdhuoreuld be to select suitable single- crosses from a diallel trial for the formation oofu dble crosses and also good combiners (high GCA) for formation of synthetics. Diallel crosses have been widely used in genestiec arrech to investigate the inheritance of important traits among sets of genotypes. These wdevr ised specifically to investigate 145 the combining ability of the parental lines for tphuerpose of identifying superior parents for use in hybrid development programmes. It isi saadbvle to carry out indirect selection for related traits which show close correlation hw yitield and exhibit high heritability because yield is considered a polygenic trait. yAsnisa lof diallel data is usually conducted according to the methods of Griffing (1956) whicahr tpition the total variation of diallel data into GCA of the parents and SCA of the cro (sYsaens and Hunt, 2002). A diallel is simple to manipulate in maize and sluiepsp important information to plant breeders about the studied populations for vargioeunse tic parameters (Vacaerto a l., 2002). In Malawi there is no known maize inbred line ori evtay that is tolerant to low pH, hence the objective of this study was to evaluate thefo prmerance of diallel crosses made with 12 maize inbred lines. Seven were from CIMMYT-Coloma,b fiour were from CIMMYT- Zimbabwe and a well-adapted line from Mala wi. 6.3 Materials and Methods 6.3.1 Experimental materials description The experimental materials comprised of seven loHw t oplerant maize inbred lines from CIMMYT-Colombia and five well adapted lines in Mawlia from CIMMYT-Zimbabwe (Table 6.1). Seed increase of the parental liness d woane at Chitala Research Station in the winter of 2010. Hand pollination by selfing and kbiunlg eight to 10 plants from each parental line was done. Diallel crosses withouitp rreoc als were made during the 2010/11 season at Chitedze Research Station in Lilongwseu imn mer and at Chitala Research Station and Bwanje Irrigation Scheme during winbtye sr tagger planting. The single crosses pedigrees have a slash separating female and maraelnet .p For example a single cross S410 has a pedigree CML144/CML288. This means that ethmea fle was parent number four which was CML144 and the male was parent numbewr h1ic0h was CML288. The male is always to the right hand side. Even in a three-cwraoys s the male is to the right hand side and the only difference is the double slashes bee fothre male. For instance CML144/CM288// CML481 implies that CML481 whicht ios the right hand side after two slashes is the male. This was adopted from CIMMYCTIM (MYT, 2012). The trial was 146 planted at four sites which included Lunyangwa lpoHw site, Tsangano low pH site, Bembeke low pH site and Chitedze optimal site. Table 6.1 Description of 12 maize parental linesd uins the diallel crosses and their origin NO. Parental line Origin Traits 1 CZL999601 CIMMYT-ZIMBABWE GLST, LBT, MSVT 2 CML481 CIMMYT-COLOMBIA Low pH 3 CML359 CIMMYT-COLOMBIA Low pH 4 CML144 CIMMYT-COLOMBIA Low pH 5 CML161 CIMMYT-COLOMBIA Low pH 6 CML172 CIMMYT-COLOMBIA Low pH 7 CML448 CIMMYT-COLOMBIA Low pH 8 CML312 CIMMYT-ZIMBABWE GLST, LBT, MSVT 9 ZL130-23 MALAWI GLST, LBT, MSVT 10 CML288 CIMMYT-COLOMBIA Low pH 11 CML202 CIMMYT-ZIMBABWE GLST, LBT, MSVT 12 CML539 CIMMYT-ZIMBABWE DT, GLST, LBT, MSVT GLST = gray leaf spot tolerant, LBT = leaf blight rtaonlet, MSVT = maize streak virus tolerant, DT = drout gh tolerant 6.3.2 Experimental procedures and design Sixty six single-cross hybrids plus two checks w learide out in a (0.1) alpha lattice design (0,1) with three replications. Plot size was twwo sro of 5.1 m per plot with 17 stations per row, one seed was planted per station and two taht ebnods of the row. 6.3.3 Description of sites The description of sites as well as managemenhte o tfr tials was given in the materials and methods section 4.3 of Chapter 4. The list of phtyepnico and agronomic traits investigated and measuring procedures are the same as in T.a1b (lCe h4apter 4). 147 6.4. Data analysis Across site analysis was performed using GenS0ta1t3 ()2. Combining ability analyses were carried out using Statistical Analysis Software S(S, A2013). Genotypes were considered fixed. Dendrogram construction was carried out CinS NS (Hintze, 2007) using UPGMA. Clustering was based on GCA effects for the inblrineeds for grain yield, Euclidean distance and standard deviation as scale type. 6.5 Results 6.5.1 Performance of diallel cros ses Mean squares for performance of diallel crosse sg rfaoirn yield and other agronomic traits are presented in Table 6.2. Across site analysr iosp ftoimal and three low pH sites indicated that all the sources of variation against the mreeads utraits were significant with the exception of the interaction of genotype for shnegl lipercentage and MSV for site. The mean performance across all four sites showed hthyabtr id CZL999601/CML144 (2.4 t ha-1) was the best performer, followed by CZL999601/C2M02L (2.1 t h-a1), CZL999601/ZL130-23 and CML481/CML288 (2.0 t -1h) a(Table 6.3). Across the three low pH sites, CML288/CML202 (1.04 t -h1)a, CML359/CML448 (1.0 t h-a1) and CML481/CML288 (0.9 t h-a1) performed relatively better than the other hysb r id (Table 6.4). At the optimal site CZL999601/CML1474. 0( t ha-1), CZL999601/CML202 (5.4 t ha-1), CML172/ZL130-23 (5.0 t h-1a) and CML481/CML288 (4.5 t h-1a) were among the top ten yielding hybrids (Table 6.5). HybridsM LC481/CML288 showed some consistency in performance across environmentse w ChZilL999601/CML144 was suited for optimal conditions. Grain yield was reduced 7b8y. 2%, ear height by 55.3%, number of ears per plant by 54.5%, grain texture by 15 .p3l%an,t height by 41.7%, 100 seed weight by 36.3%, shelling percentage by 18.8% and plagnot uvri by 62.5% (Table 6.6). Grain yield and plant vigour were the most reduced t ruanitdser low pH. Grain texture exhibited the lowest percentage reduction among the meastruarietsd. 148 6.5.2 Genetic variances, phenotypic variances earnitda bhility estimates for the diallel crosses across optimal and three low piHro ennmvents in 2011/12 Results for the diallel crosses are presented binle T 6a.7. High broad sense heritability were recorded for anthesis date (0.93), anthesis-si lkinintegrval (0.87), grain yield (0.86) and leaf blight (0.86). The lowest heritability valuea sw recorded for number of ears per plant (0.14) Table 6.2 Mean squares for diallel crosses acrpotsims aol and three low pH environments for grain yield and agronomic trani t2s0 i11/12 Source Site Genotype G x E MSE LSD CV% GY 4.78E+08** 1.47E+06** 1.07E+06** 3.50E+05 474.8 12.9 AD 34.87** 34.87** 8800** 2.866 2.716 2.15 ASI 599.97** 1.81** 1.888** 0.575 0.61 30.7 PH 496823** 460.8** 495.9* 259.4 25.8 12.9 EH 160708** 280.5** 250.8** 129.1 18.22 21.1 EPP 32.2** 0.072** 0.093** 0.204 0.31 29 RL 1322.7** 10.76** 10.4** 5.87 2.25 118 SL 21.9** 4.5** 4.1** 2.56 1.82 114 GLS 81.8* 0.99** 0.748** 0.18 0.39 25 LB 82.2** 1.49* 1.26** 0.21 0.43 26.4 MSV 0.0001 0.31** 0.42** 0.16 0.08 36.8 RUST 219.3** 0.48* 0.43** 0.15 0.36 21.1 GT 106.1** 0.88** 1.1** 0.27 0.59 25.5 SH 15036** 617.9 620.3** 478 24.8 31.4 VIG 569.76** 1.22** 1.38** 0.31 0.64 16.4 SWT 10523** 94.8** 77.2** 24.8 5.66 19 **P ≤0.01; *P≤0.05; G = genotype, E = environment, MSE = Mean sqe uearrror, LSD = Least significant difference C, V = coefficient of variation, GY = grain yieldg (hka-1), AD = days to anthesis (days), ASI = anthesis- silking interval (days), PH = plant height (cm), =E Hea r height (cm), EPP = ears per plant (#), RL =t r looodging (#), SL = stem lodging (#), GLS = gray leaf spoet adsies (1-5), LB = leaf blight disease (1-5), MSV =iz me astreak virus disease (1-5), Rust = rust disease (1-5), GgrTa i=n texture (1-5), SH = shelling percentage, V=I Gvigour (1- 5), SWT = 100 seed weight (g), # = number. 149 Table 6.3 Mean performance of diallel crosses sa corpotsimal and low pH environments in 2011/12 se ason Code GY AD ASI DS EH EPP LB GLS GT MSV PH RL RUST SWT SH SL VIG Top 10 G3 2379 80.5 1.1 82.6 56.9 0.6 1.3 1.7 2.4 1.0 81 262..2 1.8 24.4 70.9 2.4 3.0 Genotypes G10 2075 78.2 1.3 80.3 48.2 0.6 2.3 1.2 2.0 1.1 .11 192.2 2.1 24.9 74.8 1.9 3.7 G8 1984 80.2 1.4 82.5 55.0 0.7 2.3 1.6 2.0 1.0 81 255..2 1.9 27.1 73.4 1.8 2.7 G19 1919 76.3 1.7 78.7 58.8 0.6 1.3 1.9 1.8 1.0 .31 300.8 1.4 27.2 71.5 1.2 2.8 G48 1913 78.8 2.0 81.5 60.0 0.8 2.0 2.4 2.4 1.0 .31 313.2 2.3 28.1 72.8 0.8 3.5 G39 1837 79.0 1.4 81.1 57.8 0.6 3.0 2.1 2.4 1.0 .11 204.0 1.7 27.1 74.3 1.4 3.0 G2 1811 77.8 1.6 80.2 56.1 0.6 2.3 2.6 1.7 1.0 31 321..7 2.3 29.8 70.5 1.3 3.0 G16 1808 80.3 1.3 82.0 46.7 0.6 1.7 1.6 2.2 1.2 .51 058.0 1.8 27.1 69.8 1.3 2.7 G7 1657 78.3 1.9 80.7 45.7 0.6 2.7 1.8 1.9 1.1 71 214..2 1.5 24.5 70.8 2.1 3.7 G24 1631 79.1 1.9 81.8 51.4 0.9 1.3 1.9 1.9 1.7 .01 201.3 1.8 23.5 66.0 0.9 2.8 Bottom 10 Genotypes G57 965.8 79.8 1.7 82.3 59.5 0.7 1.3 2.0 1.9 1.0 0.01 3 3.5 2.3 26.8 67.1 0.4 3.4 G54 963 81.8 1.2 83.9 61.5 0.6 2.0 1.0 2.1 1.3 71 311..3 1.7 35.7 74.8 0.9 3.3 G47 943 80.2 2.1 82.5 57.6 0.7 1.0 1.8 2.3 1.0 91 242..5 2.2 25.1 75.3 0.8 3.8 G55 940.8 80.3 1.7 82.8 56.7 0.6 3.0 2.0 1.3 1.2 9.41 2 2.7 2.0 26.4 70.1 0.4 3.4 G60 842.8 78.5 1.8 81.1 55.0 0.7 1.0 1.6 1.8 1.0 0.11 3 4.5 1.8 20.1 71.8 2.6 4.6 G62 780.3 80.6 2.2 83.5 53.3 0.7 1.0 1.7 1.6 1.0 2.51 2 1.0 1.7 27.2 66.4 1.8 3.7 G41 778 80.3 1.8 82.6 46.4 0.6 4.0 1.7 2.3 1.0 61 132..3 1.7 24.8 68.5 0.4 4.0 G44 733 79.5 0.9 81.4 65.3 0.7 4.0 1.7 2.2 1.0 71 334..8 2.2 23.2 61.3 0.4 3.4 G43 711 82.0 2.2 84.9 48.1 0.7 2.0 1.7 1.7 1.1 91 230..5 1.7 28.2 65.0 1.7 3.8 G40 646 77.5 1.9 80.2 47.8 0.7 2.0 1.0 2.3 1.0 81 241..2 1.3 27.4 60.9 0.7 4.3 Mean 1366 79 2 82 54 1 1.0 2 2 1 125 3 2 26 70 1 3 LSD 478.8 1.4 0.6 1.4 9.1 0.2 1.7 0.4 0.6 0.6 12.92 .3 0.6 8.4 17.7 1.5 0.6 MSE 3.50E+05 2.9 0.6 3.2 129.1 0.0 0.4 0.2 0.3 0.22 59. 5.9 0.2 24.8 620.3 2.6 0.3 LSD = Least significant differen cMeS, E = Mean square error, GY = grain yield (kg- 1h),a AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (days), EH = ear height (cm), EPP = ears per pl(a#n),t LB = leaf blight disease, (1-5), GLS = graayf lsepot disease (1-5), GT = grain texture (1-5),V M =S maize streak virus disease (1- 5), PH = plant height (cm), RL = root lodging (R#u),s t = rust disease (1-5), SWT = 100 seed weig),h St H(g = shelling percentage, SL= stem lodgingV I(G#) ,= vigour (1-5), # = number. 150 Table 6.4 Mean performance of diallel crosses sa ctrhorsee low pH environment in 2011/2012 season Entry GY AD ASI DS EH EPP LB GLS GT MSV PH RL RUST SWT SH SL VIG Top 10 G64 1040.7 77.7 2.7 80.3 34.1 0.4 2.0 2.3 1.5 1.0 101.0 0.5 1.8 16.5 65.6 2.5 5.0 Genotypes G25 1007.7 81.9 3.4 85.3 32.6 0.8 1.7 1.3 2.2 1.0 92.7 0.5 2.0 21.8 70.3 1.3 4.7 G19 999.3 77.8 3.0 80.8 47.8 0.5 1.3 2.3 1.7 1.0 113.1 0.5 1.6 21.6 66.8 1.5 4.3 G10 970.7 79.6 2.6 82.1 40.4 0.5 2.3 1.3 1.9 1.0 104.1 0.5 2.6 20.4 69.5 2.3 5.0 G18 958.3 83.0 3.2 86.2 44.4 0.5 2.0 2.0 1.9 1.1 109.7 0.5 2.0 17.2 65.0 1.5 4.3 G6 953.0 77.4 3.0 80.4 43.8 0.5 2.7 2.1 2.6 1.0 113.4 0.5 2.3 18.9 73.6 1.0 5.0 G8 945.3 82.9 3.1 86.0 41.5 0.5 2.3 1.8 2.1 1.0 103.6 0.5 2.3 24.0 70.2 1.5 4.3 G51 896.0 80.3 3.0 83.3 45.0 0.6 3.0 2.0 2.1 1.0 100.4 0.5 2.5 20.3 65.4 1.0 4.0 G1 894.3 83.4 2.9 86.3 46.8 0.5 2.3 1.3 2.2 1.0 109.8 0.5 2.4 22.7 62.0 2.2 4.3 G20 872.0 79.3 2.9 82.2 43.6 0.6 2.7 2.1 2.1 1.0 110.5 0.5 2.3 23.0 73.4 1.2 4.3 Bottom10 G50 478.0 81.0 2.7 83.7 35.6 0.5 4.0 1.5 2.0 1.0 95.3 0.5 2.3 20.0 50.0 1.5 5.0 Genotype s G57 471.7 81.0 3.0 84.0 50.3 0.6 2.0 2.5 1.6 1.0 113.8 1.0 3.0 25.0 63.0 0.5 4.0 G47 468.3 83.7 2.7 86.3 55.0 0.6 3.0 1.8 2.2 1.0 112.2 0.5 2.8 23.8 72.8 1.0 4.0 G46 463.7 79.3 2.7 82.0 44.1 0.4 2.0 2.0 2.6 1.0 113.9 1.0 3.0 25.6 35.9 3.0 5.0 G44 458.3 80.7 2.3 83.0 46.4 0.6 2.0 2.0 2.3 1.0 105.8 0.5 2.8 20.5 57.5 0.5 4.0 G41 435.0 83.0 3.0 86.0 33.3 0.4 4.0 2.0 2.3 1.5 94.4 0.5 2.0 22.2 65.2 0.5 5.0 G56 426.3 79.3 3.7 83.0 36.0 0.5 1.0 1.8 1.9 1.0 86.7 0.5 2.0 12.7 62.5 1.0 4.0 G54 373.3 83.0 2.7 85.7 47.3 0.4 1.0 1.5 1.9 1.0 111.7 0.5 2.0 32.4 69.8 1.0 5.0 G58 360.3 85.0 2.3 87.3 38.3 0.7 3.0 1.5 2.2 1.0 102.5 1.0 2.0 17.1 61.4 1.0 5.0 G62 266.0 82.7 3.3 86.0 35.7 0.6 4.0 2.0 1.6 1.0 98.3 0.5 2.0 25.0 61.6 2.5 5.0 Mean 699.9 81.0 2.9 83.9 41.1 0.5 2.1 1.9 2.0 1.1 05.51 0.6 2.2 22.2 65.7 1.6 4.6 LSD 293.3 3.1 0.7 0.4 12.9 0.3 0.5 0.8 0.8 0.7 19.30.3 0.7 6.9 57.8 1.6 0.6 MSE 100135 3.7 0.5 4.0 61.4 0.0 0.2 0.2 0.3 0.2 . 1 450.0 0.2 18.3 1295 1.0 0.1 CV (%) 45.2 2.4 24.6 2.4 19.0 39.3 26.9 25.0 26.0 2.54 11.4 36.4 19.4 19.0 52.1 63.8 7.7 SE 316.4 1.9 0.7 2.0 7.8 0.2 0.4 0.5 0.5 0.5 12.1 .2 0 0.4 4.3 36.0 1.0 0.4 Min 266.0 77.3 2.1 80.3 28.7 0.4 1.0 1.0 1.0 1.0 .7 86 0.5 1.5 12.7 35.9 0.5 4.0 Max 1040.7 85.7 3.9 88.4 57.7 0.8 4.0 3.0 2.9 3.5 23.21 1.0 3.0 32.4 83.2 3.5 5.0 LSD = Least significant differen cMeS, E = Mean square error, CV =coefficient of variatio nS,E = error, Min = minimum, Max =m aximum,G Y = grain yield (kg h-a1), AD = days to anthesis (days), ASI = anthesis-silkinegrv ianlt (days), DS = days to silking (days), EH = h eeaigr ht (cm), EPP = ears per plant (#), LB = lelaigf hbt disease (1-5), GLS = gray leaf spot disease (1-5), GT = grain texture (1-5), MS mV a=ize streak virus disease (1-5), PH = plant hte (icgmh), RL = root lodging (#), Rust = rust dise(a1s-5e) , SWT = 100 seed weight (g), SH = shelling percentage, SL = stem lodging (IG#) ,= V vigour (1-5), # = number. 151 Table 6.5 Mean performance of diallel crossese a ot pthtimal environments in 2011/12 Entry GY AD ASI DS EH EPP GLS GT MSV PH RL RUST SWT SH SL VIG Top 10 G3 6955 74.0 0.7 74.7 89.4 1.2 1.8 2.8 1.0 173.3 3.0 1.2 42.7 81.9 4.7 1.0 genotypes G10 5386 74.0 1.0 75.0 71.3 1.0 1.0 2.3 1.2 163.8 3.7 1.0 38.6 85.3 1.0 2.3 G48 5218 74.0 2.0 76.0 85.9 1.1 1.3 2.3 1.0 180.4 5.0 1.5 37.0 80.7 1.3 2.0 G2 5210 73.0 0.3 73.3 82.7 1.0 2.7 2.3 1.0 178.1 5.7 1.8 38.0 82.7 0.3 2.0 G8 5099 72.0 0.0 72.0 95.6 1.2 1.2 1.8 1.0 191.7 8.3 1.0 36.3 83.9 2.3 1.0 G39 5026 75.0 1.3 76.3 98.1 1.0 1.3 2.7 1.0 194.8 4.0 1.0 40.7 83.8 0.3 1.0 G16 4976 76.0 0.3 76.3 70.8 1.0 1.0 2.3 1.0 124.2 1.3 1.2 38.2 79.9 1.0 1.3 G45 4713 74.0 2.0 76.0 88.2 1.1 1.0 1.8 1.0 174.1 9.7 1.2 34.0 82.3 0.0 2.3 G19 4676 72.0 0.3 72.3 92.0 1.0 1.3 2.2 1.0 181.8 5.7 1.0 43.8 86.8 0.7 1.2 G24 4640 73.3 2.7 76.0 81.1 1.3 1.3 1.8 1.0 168.9 5.0 1.0 37.5 78.3 1.7 1.0 Bottom10 G62 2323 74.3 1.7 76.0 106.2 1.1 1.0 1.5 1.0 194.7 4.7 1.0 33.8 80.6 0.3 2.3 genotypes G55 2239 77.0 0.0 77.0 92.9 1.1 1.0 2.2 1.0 186.3 2.7 1.0 31.7 75.6 0.3 2.8 G31 2142 71.0 0.0 71.0 101.0 1.0 1.7 2.7 1.0 187.7 4.3 1.0 32.5 83.1 0.7 2.3 G35 1835 76.0 0.7 76.7 95.8 1.2 1.0 1.7 1.2 174.3 10.0 1.2 33.7 74.5 0.0 2.2 G41 1805 72.0 0.3 72.3 85.5 1.0 1.0 2.5 1.2 170.7 2.3 1.2 29.8 75.1 0.3 3.0 G60 1622 69.0 1.3 70.3 84.0 1.0 1.2 2.2 1.2 172.7 10.3 1.0 27.7 75.2 0.7 4.2 G44 1556 76.0 0.7 76.7 122.2 1.1 1.0 2.0 1.0 216.9 1.0 1.0 31.5 72.5 0.3 2.8 G61 1456 74.0 0.0 74.0 104.8 1.0 1.3 1.7 1.0 190.3 8.0 1.3 24.5 77.0 0.0 4.2 G43 1381 74.0 0.7 74.7 84.3 1.0 1.0 2.2 1.0 180.2 3.0 1.0 29.5 79.4 0.0 3.5 G40 757 72.0 3.7 75.7 88.1 0.9 1.0 2.2 1.0 177.9 8.0 1.0 27.9 70.0 0.0 4.5 Mean 3450.8 73.6 1.0 74.6 91.0 1.1 1.3 2.2 1.0 11 795. .3 1.1 34.5 79.4 0.8 2.3 LSD 1649.9 0.8 1.4 1.3 328.2 0.2 0.5 0.8 0.4 38.7 .7 6 1.1 10.2 10.5 3.8 1.1 P 0.001 0.001 0 0 0.63 0.165 0.001 0.36 0.16 0.75 .0020 0.078 0.001 0.05 0.35 0.001 CV (%) 31.9 0.7 74.7 0.8 19.7 12.4 22.8 23.7 21.3 3.21 82.9 0.2 18.0 8.0 230.1 30.8 Min 757.0 69.0 0.0 70.3 70.8 0.9 1.0 1.5 1.0 1 24.12.0 1.0 24.5 70.0 0.0 1.0 Max 6955.0 77.0 3.7 77.0 122.2 1.3 2.7 2.8 1.2 .92 1610.3 1.8 43.8 86.8 4.7 4.5 LSD = Least significant differen cCeV, = coefficient of variatio nM,in = minimum, Max = maximum ,GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = anthesis- silking interval (days), DS = days to silking ()d, aEyHs = ear height (cm), EPP = ears per plant (G#L),S = gray leaf spot disease (1-5), GT = grainu treex t(1-5), MSV = maize streak virus disease (1-5), PH = plant height (cm), RLo o=t r lodging (#), Rust = rust disease (1-5), SWT00 = s 1eed weight (g), SH = shelling percentage, sStLem = lodging (#), VIG = vigour (1-5), # = number. 152 Table 6.6 Estimated percent reduction for saliehnetn potypic traits for diallel crosses under low pH versus optimal conditions Trait Low pH Optimal Percentage of optimal % Rediuocnt GY 699.9 3205.2 21.8 78.2 EH 41.1 92.0 44.7 55.3 EPP 0.5 1.1 45.5 54.5 GT 1.9 2.2 84.7 15.3 PH 105.5 180.9 58.3 41.7 SWT 22.2 34.9 63.7 36.3 SH 65.7 81.0 81.2 18.8 VIG 4.6 2.2 37.5 62.5 GY = grain yield (kg h-a1), EH = ear height (cm), EPP = ears per plant (#), =G Tg rain texture (1-5), PH = plant height (cm), SWT = 100 seed weight (g), SH = shge pllienrcentage, VIG = vigour (1-5), # = number. Table 6.7 Genetic variances, phenotypic varianncde sh aeritability estimates for the diallel crosses across optimal and three low piHro ennmvents in 2011/12 Trait Genetic varianceδ (2g ) Phenotypic variancδe2p( ) Heritability(H2b) GY 311866.67 361866.73 0.86 AD 10.67 11.53 0.93 ASI 0.41 0.47 0.87 DS 10.65 11.71 0.91 EH 50.37 65.43 0.77 EPP 0.01 0.07 0.14 LB 0.42 0.49 0.86 GLS 0.27 0.33 0.82 GT 0.2 0.27 0.74 MSV 0.05 0.11 0.45 PH 47.1 67.16 0.70 RL 1.63 1.97 0.83 Rust 0.11 0.17 0.65 SH 31.57 46.63 0.68 SL 1.96 3.03 0.65 SWT 21.33 28.39 0.75 VIG 0.30 0.36 0.83 σ2g = genotypic variancσe2, p = phenotypic variance,2 bH = broad sense heritability, GY = grain yield h(kag-1) , AD = days to anthesis (days), ASI = anthesis-sgi liknitnerval (days), DS = days to silking (days), =E He a r height (cm), EPP = ears per plant (#), LB = leaf blight, G=L Sgr ay leaf spot disease (1-5), GT = grain text1u-r5e) ,( MSV = maize streak virus disease (1-5), PH = plhaenitg ht (cm), RL = root lodging (#), Rust = rust adsisee (1-5), SH = shelling percentage, SL = stem lodging W(#T), =S 100 seed weight (g), VIG = vigour (1-5), # =m nbuer 153 6.5.3 Combining ability and inheritance Combined ANOVA results across all sites are preesde nint Table 6.8. Environment mean squares were significant for all traits measurecde epxt for maize streak virus. Genotype mean squares were significant for all traits exicnlgu droot and stem lodging as well as for grain texture and shelling percentage. GCA wasi fsicigant for grain yield, anthesis date, anthesis-silking interval, days to silking, plandt a ear height, and gray leaf spot, leaf blight, maize streak virus, rust, and plant vigour. The riancttion of GCA by site was significant for grain yield, anthesis date, root lodging, glreaayf spot, leaf blight, maize streak virus, rust, grain texture, shelling percentage and pvlaignotu r. SCA was significant for all traits measured except shelling percentage. The intenra cbteiotween SCA effects and site was again significant for all traits apart from numboef re ars per plant, stem lodging and 100 seed weight. The calculated ratios of GCA/SCA weeitrhee r higher or closer to 1.0 for the individual traits (Tables 6.8 and 6.9) suggestihneg ptresence of additive and non-additive gene action. Under low pH soil environments (Ta6b.l9e) additive gene action was predominant in the inheritance of grain yield, nuemr obf ears per plant, shelling percentage, 100 seed weight and plant vigour because the GCA /rSaCtios were greater than unity while non-additive gene action was predominant hien itnheritance of anthesis-silking interval, plant and ear height, grain texture, s atenmd root lodging, gray leaf spot disease and rust disease. However, the SCA effects wer esi gnnoitficant for shelling percentage and ears per plan t. The mean squares due to GCA and SCA were higheerr uonpdtimal than under low pH conditions and across sites (Table 6.10). Howethver m, ean squares due to GCA and SCA was higher under low pH than under optimal condnsit ioexcept for root and stem lodging. With respect to the relative contribution of GCAd a SnCA to total variation, GCA sum of squares for grain yield contributed relatively m otore variation under optimal conditions (32.2%) than across environments (28.1%) (Tabl1e) 6. .S1imilarly the sum of squares due to SCA contributed the highest amount of variatuionnd er optimal (67.8%) compared to low pH conditions and across sites. A similar tr ewnads observed for other traits. Under low pH the percentage contribution ranged from 0to.0 91.21 (Table 6.11 ). 154 Table 6.8 Combined analysis of variance for GCA SaCndA for diallel crosses for grain yield and oathgeror nomic traits across optimal and three low pH environments in 2012 Source Site Genotypes GCA GCA x Site SCA SCA x Site Error GCA/SCA ratio DF 3 65 11 33 54 162 195 GY 833.60** 2.94** 12.40** 1.86** 3.53** 1.44** 0.45 3.51 AD 345.02** 3.92** 4.36** 3.90** 4.87** 6.20** 10.16 0.90 ASI 81605.70** 8.35** 16.43** -0.70 13.31** 7.53** 3.51 1.23 DS 258820** 11.10** 109.24** -18.89 36.38** 5.66** 1.24 3.00 PH 1709.46** 1.93** 1.74** 1.77 2.13** 1.76** 279.09 0.82 EH 1132.16** 2.54** 1.76** 1.92 2.15** 1.89** 132.08 0.82 EPP 38.90** 1.58** 1.66 -7.40 1.71** -18.16 89.27 .970 RL 118.80** 1.24 1.23 1.83** 1.89** 1.46** 7.65 05.6 SL 44.70** 1.27 1.61 1.69 1.17* 1.10 2.00 1.38 GLS 429.14** 4.91** 5.72** 5.49** 5.38** 3.62** 0.19 1.06 LB 251.47** 3.76** 3.86** 4.88** 3.70** 3.70** 0.32 1.04 MSV 0.02 1.72** 2.17** 3.94** 1.85** 2.58** 0.015 .117 RUST 1394.02** 3.09** 3.76** 2.12** 3.30** 2.94** 0.15*** 1.14 GT 28.55** 1.01 1.05 1.76** 1.46* 1.47** 3.92 0.72 SH 36.10** 1.29 1.61 2.16** 1.17 1.35** 599.50 1 .38 SWT 277.60** 1.68** 1.12 0.450 2.30** 0.48 44.76 409. VIG 1316.1** 2.38** 1.93* 3.19** 2.91** 3.24** 0.43 0.66 ***P ≤0.001; **P≤0.01; *P≤0.05; GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daPyHs) =, plant height (cm), EH = ear height (cm), EPP = ears per plant (#), RLo =o tr lodging (#), SL = stem lodging (#),GLS = glerayf spot disease (1-5), LB = leaf blight diseMaSsVe, = maize streak virus disease (1-5), Rust = rust disease (1-5), GT = grain teex t(u1r-5), SH = shelling percentage, SWT = 100 seeeidg hwt (g), VIG = vigour (1-5), # = number. 155 Table 6.9 Combined analysis of variance for GCA SaCndA for diallel crosses for grain yield and oathgeror nomic traits across three low pH environments in 2012 Source Site Genotypes GCA GCA x Site SCA SCA x Site Error GCA/SCA Ratio DF 2 65 11 22 54 108 GY 223.70** 0.48** 1.09** 0.32** 0.57* 0.05* 0.18 .192 AD 573.40** 43.88** 45.70** 53.50** 44.30** 89.30** 11.34 1.03 ASI 379684.90** 36.59** 16.00** 13.95** 44.20** 384.0** 3.65 0.36 DS 471947.80** 28.80** 52.70** 29.90** 34.98** 340.4** 7.44 1.51 PH 331735.00** 601.70** 154.57 686.00** 686.70** 35.090** 172.90 0.23 EH 49538.45** 367.60** 307.40** 150.70** 379.70** 023.60** 62.37 0.81 EPP 159.00** 35.70 40.00 54.00* 34.90 41.45 32.68 .15 1 RL 44.40** 2.94** 2.16** 0.97 3.15** 1.50** 7.70 06.9 SL 172.70** 0.56** 0.18 0.27 0.64** 0.31** 2.00 08.2 GLS 115.10** 0.92** 0.80* 0.76** 0.96** 0.47** 0.2 1 0.83 LB 114.96** 1.036** 1.23** 1.32** 1.01** 0.68** 0.33 1.22 MSV 0.20** 0.17** 0.28** 0.24** 0.15** 0.19** 0.03 1.87 Rust 317.50** 2.50** 1.60 2.40* 2.70** 1.90* 1.40 .509 GT 152.00** 0.99** 0.80** 0.60** 1.06** 0.60** 0.49 0.75 SH 2679.30* 877.93 851.16 -315.60 839.61 -357.79 9.2704 1.01 SWT 5964.70** 76.39** 120.42** -50.39 74.15** -210.6 22.39 1.62 VIG 429.10** 2.16** 61.23** 1.97** 45.39* 1.98** 18. 0 1.35 **P ≤0.01; *P≤0.05; GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daPyHs) =, plant height (cm), EH = ear height (cm), EPP = ears per plant (#), RL = roodtg liong (#), SL = stem lodging (#), GLS = gray lepaoft sdisease (1-5), LB = leaf blight disease, MSmVa i=z e streak virus disease (1-5), Rust = rust disease (1-5), GT = grain texture (,1 S-5H) = shelling percentage, SWT = 100 seed wegig),h Vt I(G = vigour (1-5), # = number. 156 Table 6.10 Mean squares for GCA and SCA effectse ur nddifferent environmen ts Trait Across Optimal Low pH GCA SCA GCA SCA GCA SCA GY 6.19** 1.76** 2423935.00** 1039389.32** 1.09** 0.57* AD 43.63** 49.41** 6.23** 6.96** 45.70** 44.30** ASI 57.76** 46.70** 0.71** 0.82** 16.00** 44.20** DS 135.50 45.13** 7.77** 6.48** 52.7** 34.98** EH 234.33 284.98** 107.00 100.53 154.57 34.90 EPP 148.24 152.50** 0.01 0.01 40.00 34.90 LB 1.25** 1.51** 0.49 0.73** 1.23** 1.01** GLS 1.08** 1.02** 0.30** 0.17** 0.80* 0.96** GT 4.11 5.74** 0.11 0.10 0.99** 1.06** MSV 0.33** 0.28** 0.02 0.02 0.28** 0.15** PH 488.42 596.17** 190.78 159.43 601.70** 686.0 0** RL 9.41 14.46** 11.23* 10.26** 2.16** 3.15** RUST 0.58** 0.51 0.02 0.03 2.50** 2.70** SWT 50.18 103.27** 18.40 39.90 120.40** 74.2** SH 964.10 704.60** 32.96** 17.05 851.16 839.61 SL 3.24 2.34 3.05 1.82 0.18 0.64** VIG 0.84** 1.26** 0.68 0.72** 61.23** 45.39** **P ≤0.01; *P≤0.05; GSA = general combining ability, SCA = spice ciofmbining ability, GY = grain yield (kg- 1h),a AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (days), EH = ear height (cEmP)P, = ears per plant (#), LB = leaf blight dise, aGsLeS = gray leaf spot disease (1-5), GT = graxintu tree (1-5), MSV = maize streak virus disease (1-5), PH = plant height (cm), RLo o=t r lodging (#), Rust = rust disease (1-5), SWT00 = s 1eed weight (g), SH = shelling percentage, sStLem = lodging (#), VIG = vigour (1-5), # = number . 157 Table 6.11 Relative percent contribution of sumsq oufa res for GCA and SCA to total sum of squareoss asc ernvironments Sum of squares Percentage of Percentage of Trait Environment GCA Total SS total SS SCA Total SS total SS GY Across 68.10 242.00 28.10 95.10 242.00 39.30 Optimal 26663289.0 0 82760312.2 0 32.20 56127023.0 0 82760312.2 0 67.80 Low pH 1.09 54.96 2.00 0.57 54.96 1.03 EH Across 2577.70 67027.00 3.80 15389.00 67027.00 23.00 Optimal 1177.20 6605.90 17.80 5428.70 6605.90 82.20 Low pH 307.40 8915.74 3.44 379.70 8915.74 4.30 EPP Across 1630.60 6680.30 24.40 8236.20 6680.30 123.30 Optimal 0.10 0.50 17.40 0.40 0.50 82.80 Low pH 40.00 8003.60 0.45 34.90 8003.60 0.01 GT Across 45.20 1519.30 3.00 309.80 1519.30 20.40 Optimal 1.20 6.40 19.30 5.20 6.40 80.70 Low pH 0.80 8766.00 0.50 1.06 8766.00 0.01 PH Across 5372.60 133576.10 4.00 32192.90 133576.10 24.10 Optimal 2098.60 10707.80 19.60 8609.30 10707.8.00 80.40 Low pH 154.57 108297.00 0.14 686.70 108297.00 0.63 SWT Across 552.00 10894.20 5.10 5576.40 10894.20 51.20 Optimal 202.40 2357.20 8.60 2154.80 2357.20 91.40 Low pH 120.42 8770.00 1.37 74.15 8770.00 0.85 SH Across 10605.10 222803.70 4.80 38049.10 222803.70 0.40 Optimal 362.50 1282.90 28.30 920.40 1282.9 71.70 Low pH 851.16 9116.80 9.34 839.61 9116.8 9.210 VIG Across 9.20 168.20 5.50 68.00 168.20 40.40 Optimal 7.50 46.20 16.20 38.70 46.20 83.80 Low pH 61.23 5597.50 1.09 45.39 5597.50 0.81 GY = grain yield (kg h-a1), EH = ear height (cm), EPP = ears per plant (G#T), = grain texture (1-5), PH = plant height (cmW), TS = 100 seed weight (g), SH = shelling percen tage, VIG = vigour (1-5). 158 Twelve maize inbred lines were clustered basedh eo nG tCA effects for grain yield. Two main clusters were observed (Figure 6.1) with a highh ceonpetic correlation ofc orp = 0.87. The pattern mostly followed the origin of the maize riendb lines such that six out of seven inbred lines from CIMMYT-Colombia (tropical) were groupeind the second cluster. The first cluster comprised of inbred lines with low GCA for graine lydi while the second cluster was comprised of inbred lines of medium GCA and therd t hciluster had CZL999601 which had the highest GCA for grain yield. Inbred_line CML312 CML161 CML539 ZL130-23 CML448 CML202 CML172 CML359 CML288 CML144 CML 481 CZL999601 2.50 2.08 1.67 1.25 0.83 0.42 0.00 Dissimilarity Figure 6.1 Dendrogram of 12 maize inbred lines db aosne GCA effects for grain yield across four environments for 2011/12 season 159 6.5.3.1 Estimated general combining ability ef feocr t1s2 inbred lines for grain yield and agronomic traits across low pH and optimalr oenmvients in 2011/12 A positive, high and significant GCA effect for lydie was recorded for inbred line CZL999601 (0.55) (Table 6.12). Also positive and high but sniogtnificant values were observed for grain yield for inbred lines by CML481 (0.18) followed bCyML144 (0.13) and CML 288 (0.13) and significant negative GCA was recorded for indb lriene CML161 (-0.38). For days to anthesis positive and significant GCA values webrese orved for lines ZL130-23 (1.2) and CML312 (1.15). Negative and highly significant GCwAa s recorded for inbred line CML288 (-0.73). Selecting for earliness to maturity sho bueld done by selecting inbred lines with low and negative GCA effects. Positive and high valfuoer sG CA effects were observed for anthesis-silking interval for inbred lines CML2818. 0(6) and CML481 (0.75). Negative and lowest GCA values were recorded for inbred lines LC1M61 (-1.1) and CML172 (-0.9). For stress tolerance breeding, lines with negative GfoCr Ath is trait would be preferred. Positive and high GCA effects were observed for bneurm of ears per plant in lines CZL999601 (1.26), CML481 (1.24), ZL130-23 (1.12) and CML2818.0 (8). When selecting for prolificacy, lines with high and positive GCA for this trait wlodu be preferred. Positive and high GCA effects were observed for grain texture for inblrineeds ZL130-23 (0.36) followed by CML312 (0.17). When selecting for flint and semi-flintinse, slines with negative GCA should be selected since a score of 1 is better than a socf o5r ein 1-5 scale. In terms of diseases, positive and high GCA effects were observed for gray leaoft sfopr lines CML172 (0.15), CML288 (0.13) and CML202 (0.11). Inbred line CML448 (-0). 1h6ad negative and significant GCA. Selecting for tolerance to this disease should obnee dby selecting lines with negative GCA. Positive and significant GCA effects were obserfvoer dM SV in lines CML288 (0.10). Positive GCAs values were also observed for inbred lines C16M1L (0.08) and CML35 (0.08). The lowest and negative GCA was observed for lines (C20M2L (-0.08), CZL999601 (-0.06) and CML539 (-0.04). Selecting lines for tolerance tois tdhisease could be achieved by selection lines with negative GCA. For leaf blight diseasieg,h h and positive GCA’s were recorded for inbred lines CML172 (1.7) followed by CML481 (1.a2n) d the lowest GCA was recorded for line CML539 (-3.3) . 160 Table 6.12 Estimated general combining abilityc etsff feor 12 inbred lines for grain yield and agroinco tmraits across low pH and optimal environments in 2011/12 Line Name GY AD ASI DS EH EPP GLS GT LB MSV PH RL uRst SWT SH SL VIG G1 CZL999601 0.55** -0.21 -0.17 -0.53 -0.43 1.26 0.05 -0.13 -0.4 -0.06 0.71 0.18 0.04 0.68 5.19 0.22 -0.04 G2 CML481 0.18 -0.26 0.75 -1.26 1.21 1.24 -0.02 -0.24 1.2 -0.04 1.25 0.39 -0.08 -0.25 -3.31 0.38* -0.2 G3 CML359 0.04 -0.63 0.6 -0.18 0.44 0.43 -0.03 -0.05 0.4 0.08 3.42 -0.21 -0.03 -0.48 -4.12 0.23 -0.06 G4 CML144 0.13 -0.32 0.69 -1.25 1.04 0.29 0.03 -0.28 1.0 -0.02 1.19 -0.36 -0.06 0.96 -0.34 0.02 -0.05 G5 CML161 -0.38** 0.12 -1.1 0.93 0.69 -1.21 0.01 0.22 0.7 0.08 -1.62 -0.23 -0.06 -0.51 -0.18 -0.16 0.16 G6 CML172 -0.02 -0.13 -0.91 1.06 1.69 -1.19 0.15 -0.09 1.7 0.03 -0.58 0.33 0.09 0.56 6.27* -0.02 -0.06 G7 CML448 -0.15 -0.27 -0.29 -0.08 -1.73 0.48 -0.16* -0.12 -1.7 -0.04 -2.3 -0.41 -0.04 1.1 -1.81 -0.16 0.05 G8 CML312 -0.28 1.15* 0.45 0.67 -0.38 -2.29 0.00 0.17 -0.4 0.00 -0.16 0.44 0.1 -0.58 1.34 0.07 0.24 G9 ZL130-23 -0.13 1.20* -0.87 2.01 1.11 1.12 -0.02 0.36 1.1 -0.01 -0.45 -0.06 0.09 -0.45 -0.63 -0.18 0.02 G10 CML288 0.13 -0.73** 1.06 -1.64 -0.9 1.08 0.13 -0.07 -0.9 0.10* 0.97 0.26 -0.15 0.68 -1.6 -0.11 0.07 G11 CML202 0.01 -0.29 0.04 -0.47 0.56 -0.2 0.11 0.05 0.6 -0.08 1.97 -0.43 0.05 -0.62 -1.46 -0.16 -0.01 G12 CML539 -0.07 0.39 -0.24 0.74 -3.31 -1.01 -0.23 0.16 -3.3 -0.04 -45 0.09 0.04 -1.09 0.65 -0.14 -0.11 **P ≤0.01; *P≤0.05; GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daEyHs) =, ear height (cm), EPP = ears per plant (#), GLS = gray leaf spot disease (1-5), G gTr a=in texture (1-5), LB = leaf blight disease, M =S mVaize streak virus disease (1-5), PH = plangt hte (icm), RL = root lodging (#), Rust = rust disease (1-5), SWT = 100 seed weig),h St H(g = shelling percentage, SL = stem lodginVgI G(# )= vigour (1-5), # = numbe r. 161 Similarly, selecting for tolerance to this diseacsoeu ld be done by selecting lines with negative GCA effects. With respect to rust, poesi taivnd high GCA effects were observed for lines CML312 (0.1), CML172 (0.09) and ZL130-2(03. 09). The lowest and negative GCA was observed for line CM288 (-0.15). Selectliinge s for tolerance to this disease would be achieved by selecting lines with negaGtivCeA . High GCA effects were observed for plant heighitn ibnr ed lines CML359 (3.42), CML202 (1.97), followed by CML481 (1.25). The negative alonwdest GCA was observed for line CML539 (-45.0). For ear height a high GCA effectse rew observed for inbred lines CML172 (1.69) followed by CML481 (1.21). Negativned a lowest GCA was observed for line CML539 (-3.31). Where tall varieties are prerefed, lines with high and positive GCA could be preferred. Positive and significant vafloure G CA effects was observed for stem lodging in inbred lines CML481 (0.38), inbred liCneM L359 (0.23) also had a high value. The lowest and negative GCA values were recordre dlin feos ZL130-23 (-0.18) followed by CML448, CML202 and CML161 all with a value of. 1-06. High GCA effects were observed for root lodging for CML312 (0.44), CML4 8(01.39) and CML172 (0.33). The lowest and negative GCA were recorded for lines C2M02L (-0.43) and CML448 (-0.41). When selecting for tolerance for both root lodgainngd stem lodging, lines with negative GCA would be preferred. High GCA effects were obvesedr for plant vigour for inbred lines CML312 (0.24), CM161 (0.16), and negative and lotw GeCsA was recorded for CML539 (-0.11). In terms of 100 seed weight, high GCA cetfsfe were observed for inbred lines CML448 (1.1), CML144 (0.96) and CML288 (0.68). Tlhoew est and negative GCA values were observed for lines CML539 (-0.09), CML202 6(-20).. With respect to shelling percentage, high GCA effects were observed fore idn blirne CML172 (6.27), followed by CZL999601 (5.19). The lowest and negative GCA wbasse orved for lines CML359 (-4.12) and CML481 (-3.3). For plant vigour, 100 seed wet,i gahnd shelling percentage, inbred lines with higher and positive GCAs would be prreefedr. 162 6.5.3.2 Estimated specific combining ability esf ffeocrt 12 inbred lines for grain yield and agronomic traits across low pH and optimalr oenmvients in 2011/12 Results for the estimated SCA effects in the dli aclrloesses are presented in Appendix 12. The promising specific combinations for grain y iewldere CZL999601/CML144 (0.94), CML144/CML202 (0.73) and CML481/CML288 (0.70). Thpeo orest combinations were CZL130-23/CML202 (-0.65) and CZL999601/CML481 (-50)..6 When breeding for high yield, positive SCA values are desirable. For dtao y5s0 % pollen shedding, the best specific combinations were CML161/CML288 (4.03), CML144/ZL01-233 (3.8), followed by CML312/CML288 (3.09). The poor combinations were LC3M59/CM288 (-6.71), CML202/CML539 (-3.7) and CML161/ZL130-23 (-2.56).e lSecting for earliness to maturity should be done by selecting inbred linneds haybrids with low and negative GCA and SCA effects, respectively. The highest SCAc etsff ewere observed for anthesis-silking interval for crosses and the best combinations wCerMeL161/CML288 (4.14), CML359/CML288 (3.9) and CML144/CML539 (3.39). LowC SA values were recorded for the following hybrids: CML288/CML539 (-3.55), LZ130-23/CML288 (-3.5), CML288/CML202 (-3.12) and CML144/CML161 (-2.9). F sotress tolerance, hybrids with negative SCA for this trait would be preferre d. Positive and high SCA effects were observed for bneurm of ears per plant for hybrids CML448/CML288 (16.6) CML144/CML202 (7.4) and ZL13203-/CML288 (6.37) and the poor combinations were CZL999601/CML481 (-5.37),1 Z3L0-23/CML288 (-4.91) and CML481/CML448 (-4.51). When selecting for prolificcya, lines and hybrids with high and positive GCA and SCA for this trait would be prerefedr. Significant and high SCA effects were observed for grain texture for hybrids ZL1330/C-2ML539 (2.75), CML161/CML202 (2.51) followed by CML312/ZL130-23 (2.19). The poeostr combinations were ZL130- 23/CML202 (-0.95), CML481/ZL130-23 (-0.84) and CM6L11/CML539 (-0.83) but when selecting for flint and semi-flintiness, lines ahnydb rids with negative GCA and SCA, respectively, could be selected. In terms of diesse,a psositive and high GCA effects were observed for gray leaf spot for hybrids CZL99960M1/LC359 (0.86), CM172/ZL130-23 (0.64), and CML144/CML202 (0.62). Hybrids CZL9996/C01ML202 (-0.61), CML161/CML448 (-0.52) and CZL999601/CML481 (-0.4h8a)d negative SCA. Selecting for tolerance to this disease could be done byc tsinegle lines and hybrids with negative 163 GCA and SCA, respectively. Positive and high SCfAe cetsf were observed for MSV in hybrids CML359/CML288 (0.92), CML161/CML172 (0.47a) nd CML172/ZL130- 23(0.23). The lowest and negative SCA was obserfover dC ML161/CML288 (-0.27), CML359/CML161 (-0.25), and CML359/CML172 (-0.20)e. lSecting hybrids for tolerance to this disease could be achieved by selectionlin tehse and hybrids with negative GCA and SCA, respectively. For leaf blight disease, highd apnositive SCA’s were recorded for hybrids CML481/CML144 (0.88) and CML172/CML202 (03.)8 followed by CML172/CML312 (0.77) and the lowest SCA’s were rredceod for hybrids CML312/CML202 (-0.6), CML481/CML288 (-0.6) and ZL01-323/CML288 (-0.59). Similarly, selecting for tolerance to this diseacsoeu ld be done by selecting lines and hybrids with negative GCA and SCA effects, respveclyti. With respect to rust disease, positive and high SCA effects were observed in ihdysb rCML172/CML448 (0.51), CZL999601/CML359 (0.44) and CML202/CML539 (0.42)h. eT lowest and negative SCA’s were observed for hybrid CML448/ZL130-23 4(-90)., CZL999601/CML312 (- 0.46) and CML144/CML202 (-0.44). Selecting linesr tfolerance to this disease could be achieved by selection hybrids with negative SCA. Positive and high SCA effects were observed fonr tp hlaeight in hybrids CML144/CML202 (18.64), CML448/CML288 (13.18), followed by CML44C8M/ L202 (12.83). Negative and lowest SCA’s were observed for hybrid CML481/CML4 (4-816.84), CML202/CML539 (-13.13) and CML144/ZL130-23 (-13.03). For ear het,i gpositive and high SCA effects were observed for hybrids CML448/CML288 (12.49)l ofowled by CML448/CML202 (8.12) and CML172/CML288 (7.96). Negative and lotw SeCsAs were observed for hybrids CML161/CML312 (-8.88), CML288/CML539 (-8.39) and CLM202/CML539 (-8.15). Where tall varieties are preferred, lines and hdysb rwi ith high and positive GCA and SCA could be preferred. In terms of lodging, positivned ahigh SCA effects were observed for stem lodging in hybrids CM481/CML359 (2.38) follodw eby CML202/CML539 (1.24) and CZL999601/CML144 (1.20). The lowest and negea StivCA were recorded for hybrids CZL999601/CML481 (-0.72), CZL999601/CML359 (-0.68a)n d CML481/CML288 (-0.61). Positive and high SCA effects were obsde rvfeor root lodging for CML172/CML448 (3.29), CML481/CML448 (2.82) and CM4L41/CML288 (2.35). The lowest and negative SCA were recorded for hybridMs L1C72/CML539 (-2.60), CML161/CML448 (-0.70) and CML481/CML228 (-1.60). Wenh selecting for tolerance 164 for both root lodging and stem lodging, inbred sli naend hybrids with negative GCA and SCA, respectively could be preferr ed. Positive and high SCA effects were observed forn tp lvaigour for hybrids, Z130- 23/CML288 (1.08), CML312/CML539 (1.04), CZL999601M/CL448 (0.99) and negative and lowest SCA’s were recorded for hybrids CZL9919/6Z0L130-23 (-0.70), CML172/CML481 (-0.60) and CML481/ZL130-23 (-060n). tIerms of 100 seed weight, positive and high SCA effects were observed forr ihdy Cb ML288/CML539 (9.44). Also high and positive values were observed for inbriende sl CML312/CML202 (5.84), CML359/CML161 (5.57), and CML448/ZL130-23 (5.04)h. eT lowest and negative SCA were observed for hybrids ZL130-23/CML539 (-6.0C5)M, L448/CML539 (-4.86) and CML161/CML202 (-4.81). With respect to shelling cpenrtage, positive and high SCA effects were observed for hybrids, CZL999601/CML 17(421.02) followed by CML481/CML144 (10.49) and CML288/CML539 (9.86). T lhoewest and negative SCAs were observed for hybrids CML172/CML448 (-15.15)M, LC448/CML288 (-12.54) and CZL999601/CML539 (-10.72). For plant vigour, 100e ds eweight, and s1 helling percentage inbred lines and hybrids with higher apnodsitive GCA’s and SCA’s, respectively, would be preferre d. 6.5.4 Pearson’s correlation coefficients for dl icarlolesses between grain yield and other agronomic traits at optimal environment Pearson correlation was carried out to study thsoec aiastions between grain yield and some agronomic traits as well as among themselves. Tehsuel trs are presented in Table 6.13. Grain yield was positively and significantly corarteeld with anthesis date (0.23), days to 50% silking (0.22) and highly significantly corretelad with 100 seed weight (0.31). It was highly significantly negatively correlated with lfe balight disease (r = -0.29). In terms of diseases, gray leaf spot was positively and higshiglyn ificantly correlated with rust (r = 0.34). Plant vigour showed positive and signific aconrtrelation with leaf blight disease (0.40). Days to 50% pollen shed (days to anthewsais) highly significantly and positively correlated with 100 seed weight across optima l 0(.r2 =7). 165 Table 6.13 Pearson’s correlations coefficientsr uonpdtiemal conditions GY AD ASI DS PH EH RL SL EPP SH GT LB GLS Rust MSV VIG AD 0.23* ASI -0.01 -0.16 DS 0.22* 0.90** 0.25* PH -0.19* -0.07 0.08 -0.04 EH -0.11 -0.03 0.01 -0.03 0.80** RL -0.03 -0.02 0.03 0.00 -0.04 -0.1 SL 0.05 -0.01 -0.08 -0.04 -0.12 -0.1 -0.13 EPP -0.14 -0.01 -0.07 -0.04 -0.06 -0.06 0.03 0.05 SH 0.20 -0.26** -0.03 -0.27** -0.13 -0.12 0.12 0.00 -0.06 GT -0.10 -0.01 -0.04 -0.03 -0.01 -0.01 -0.12 0.05 -0.01 0.02 LB -0.29** -0.11 -0.03 -0.12 0.07 0.08 0.19 0.17 0.02 0.09 -0.01 GLS 0.10 -0.26** 0.01 -0.25 0.01 -0.05 0.08 0.00 -0.06 0.16 0.23 -0.06 Rust 0.06 0.02 0.00 0.02 0.04 0.00 0.03 0.23 0.01 0.11 0.11 -0.05 0.34** MSV -0.18 0.11 -0.06 0.09 -0.05 -0.12 0.16 -0.04 0.04 0.02 -0.01 0.21 -0.05 -0.06 VIG -0.49** -0.14 -0.02 -0.14 0.12 0.04 0.06 0.06 0.15 -0.09 0.15 0.40** -0.22 0.00 0.24 SWT 0.31** 0.27** -0.06 0.24 -0.06 -0.05 -0.21 0.11 0.16 -0.03 0.07 -0.24 -0.02 -0.02 -0.05 -0.29** **P ≤0.01; GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daPyHs) =, plant height (cm), EH= ear height (cm), RL = root lodging (#), SL= stem lodging (#), EPP =s epaerr plant (#), GLS = gray leaf spot disease ,( 1G-T5 )= grain texture (1-5), LB = leaf blight dise,a Ms SV = maize streak virus disease (1-5), Rust = rust disease (1-5), SWT = s 1e0e0d weight (g), SH = shelling percentage, VIGig o=u vr (1-5), # = number. 166 6.5.5 Pearson’s correlation coefficients for dl icarlolesses between grain yield and other agronomic traits at low pH environments Results are given in Table 6.14. Grain yield wassi tpivoely and significantly correlated with shelling percentage (r = 0.28), 100 seed weig=h t0 (.r5 7), plant height (r = 0.81), ear height (r = 0.64), stem lodging (r = 0.35) and grain terex tu(r = 0.34). GY was negatively but significantly correlated with anthesis date (r =.3 -10), anthesis-silking interval (r = 0.70), root lodging (r = -0.32), and plant vigour (r =4 -60).. With respect to plant vigour, the lower the score the better the plant vigour while theh ehrig the grain weight the higher the yield. So with correlation analysis this is negative aisastoiocn while in real sense genotypes with high plant vigor (low score) gives higher yield a tnhdis is a positive association. Correlation among other traits showed that plant height warse claotred with ear height (r = 0.85), 100 seed weight (r = 0.59) but was negatively and sfiicgannitly correlated with plant vigour (r = -0.46), root lodging (r = -0.26). Grain texturea sw positively and significantly correlated with shelling percentage (r = 0.36), 100 seed wte (i0g.h3). It was negatively correlated with plant vigour (r = -0.52), and rust disease (r =3 7-0).. Shelling percentage was positively and significantly correlated with 100 seed weight (r 0=.5 4), but was negatively and significantly correlated with root lodging (r = 3-00.), rust disease (r = -0.23), and plant vigour (r = -0.4) . 6.6 Discussion Effective selection of inbred lines for the prodiounct of maize hybrids requires prior information of the inbred linep er se and the behaviour of the line in a particular hidy br combination. Malawi has a humid sub-tropical cliem, awt hile Colombia has a tropical climate. In this study, both inter and intra-crcoosms binations from tropical and sub-tropical maize inbred lines indicated good yield performa unncdeer optimal condition and across sites. The top yielding hybrid was a cross betwseuebn- tropical and tropical inbred lines. Under low pH conditions, the top yielding hybrid sw a cross involving tropical inbred lines, suggesting that tropical germplasm was mt olreer ant to low pH than sub-tropical germplasm. The results were similar to what waos rrtepd by Magnavaceat al. (1987) who reported that Brazilian maize inbred lines (tropl)i cwaere generally mor e 167 Table 6.14 Pearson’s correlation coefficients eavnedl ol f significance under low pH for diallel cerso ss GY AD ASI DS PH EH RL SL GT LB GLS Rust MSV VIG SWT EPP AD -0.31** ASI -0.70** 0.25** DS -0.48** 0.96** 0.50** PH 0.81** -0.20** -0.68** -0.35** EH 0.64** -0.10** -0.47** -0.25** 0.85** RL -0.32** 0.13** 0.56** 0.27** -0.26** -0.02 SL 0.35** -0.09 -0.05 -0.10* 0.34** 0.33** 0.33** GT 0.34** -0.09 -0.39** -0.19** 0.25** 0.07 -0.47* * -0.16** LB 0.66** -0.20** -0.42** -0.26** 0.63** 0.60** 0.12** 0.47** 0.09 GLS 0.64** -0.20** -0.36** -0.25** 0.65** 0.69** 0.19** 0.42** 0.03 0.77** Rust -0.05 0.06 0.38** 0.16** -0.04 0.22** 0.79** .308** -0.37** 0.33** 0.41** MSV 0.34** -0.06 -0.04 -0.06 0.30** 0.42** 0.40** .039** -0.11 0.57** 0.64** 0.60** VIG -0.46** 0.13** 0.69** 0.31** -0.46** -0.15** 0.86** 0.23** -0.52** -0.03 0.04 0.82** 0.36** SWT 0.57** -0.20** -0.50** -0.28** 0.59** 0.45** -0.23** 0.19** 0.30** 0.43** 0.44** -0.10 0.20** -0.40** EPP 0.04 0.40** 0.01 0.36** 0.16** 0.14** -0.01 05. 0 0.07 0.00 0.04 0.01 0.02 -0.08 -0.02 SH 0.28** -0.03 -0.28** -0.10* 0.29** 0.19** -0.30** 0.01 0.36** 0.11 0.08 -0.23** -0.03 -0.40** 0.54* * 0.14** **P ≤0.01; GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daPyHs) =, plant height (cm), EH= ear height (cm), RL = root lodging (#), SL= stem lodging (#), EPP =s epaerr plant (#), GLS = gray leaf spot disease ,( G1-T5 )= grain texture (1-5), LB = leaf blight dise,a MsSV = maize streak virus disease (1-5), Rust = rust disease (1-5), SWT = 100 seeidg hwt e(g), SH = shelling percentage, VIG = vigou-5r )(,1 # = number. 168 tolerant than USA inbred lines to Al effects. Inrm tes of grain yield for the hybridpse r se, the results were similar to what was reported brye zP é(2008). Maize yields were greatly reduced across low pHes s (i7t 8.2%), followed by plant vigour (62.5%) and ear height (55.3%). The results foirn g yraield reduction were consistent with findings from other researchers. Welckeet ra l. (2005) reported that soil acidity reduces maize yields by up to 70% on 8 million hectares in devpeinlog countries and that on these soils maize yield is reduced due to Al or Mn totxyi caind Ca, Mg, P and Mo deficiencies (Aldrich et al., 1975; Granadoest al, 1993) . The results for combining ability and inheritanctued sies showed that both additive and non-additive gene action were involved and thicso inss istent with what was reported by Badawy (2013) who indicated that the ratio of GCnAd aSCA revealed the presence of additive and non-additive types of gene action. cTohnecept of the ratio of GCA/SCA was described by Baker (1978) that when this ratio oisr em than a unit, there is preponderance of additive gene action, but if it is less thann ait uthe preponderance is towards non-additive gene action. However, the results were not connsti swteith what was reported by Iqbeatl al. (2007) who indicated that none of the cross comatbioins exhibited desirable significant SCA effects for all the characters measured. Isn stthuidy higher GCA than SCA effects for grain yield were observed in both sub-tropical atrnodp ical inbred lines CZL999601 and CML481, respectively. In addition these two lineasd hgood GCA effects for other traits, for instance, CZL999601 was also a good generalb cinoemr for number of ears per plant and shelling percentage while CML481 was also ad ggoeoneral combiner for traits such as ear height, ears per plant and plant height.l eW fohri other traits some inbred lines showed higher GCA effects for plant vigour (CML3,1 120) 0 seed weight (CML448) and shelling percentage (CML172), gray leaf spot (CM2L)2 a0s well as anthesis dates (ZL130- 23) which were either sub-tropical in the case oMf LC312, CML202 and ZL130-23 or tropical inbred lines in the case of CML448 and C1M72L. This suggested the breeding of DC hybrids as an option to combine a number of irmtapnot traits for low pH toleranc e. At a cut of point of 1.0, the UPGMA clustered thneb ried lines into two main clusters through use of GCA effects for the inbred lines g forarin yield with Euclidean distance and standard deviation as scale type. At a high copthice cnoerrelation of crop = 0.87, the pattern 169 mostly followed the origin of the maize inbred lsin seuch that six out of seven inbred lines from CIMMYT-Colombia (tropical) were grouped in t hsecond cluster. Srdeict al. (2007) reported similar results, that cluster analysisn gu sSiCA effects was in good agreement with the origin of ten inbred lines, and was very us einfu lconfirming predicted heterotic patterns . With respect to SCA, the higher specific combinnasti ofor yield, plant vigour and 100 seed weight were observed in inter-crosses; CZL999601L/C14M4 and CML144/CML202. The former had a negative SCA for gray leaf spot dies,e astsem lodging implying that it was tolerant while the latter was also a good speciofimc bination for plant height, ear height and number of ears per plant. This also suggehsete bdr teeding of DC hybrids as an option to combine a number of important traits for low ptoHle rance which could broaden the genetic base . It was found that under stress the relationshipw ebent characteristics is affected by the environment at phenotypic level (Chaubey and Si n1g9h9,4; Ojo et al., 2006). In this study, grain yield at optimal environment was positivenlyd a significantly correlated with anthesis date, shelling percentage, stem lodging and 10d0 wseeight. Late maturing genotypes have adequate time to accumulate carbohydrates, leatod ihnig h yields. Shelling percentage and 100 seed weight are yield related traits suchh tihgaht yielding genotypes tend to show high values for these traits. These results were coennsti wstith what was reported by Chinnadurai and Nagarajan (2011), while under low pH grain dy iwelas negatively correlated with anthesis date. The results under low pH were ctoenst iswith findings from other researchers (Magorokosho et al., 2003; Kaoentg aal ., 2007; Ndhela, 2012) who reported negative correlation under drought stress. Underer ss t silk emergence is delayed, increasing the anthesis-silking interval (Kaongda Nanhlane, 2004). In the present study, it could be due to cold stress associated with lows ipteHs so that the plants require a higher number of days to attain the required heat unitasy (degree concept) which delays flowering but is not related to the amount of cahrybdorate accumulation in the grain as is the case with late maturing maize genotypes. Grain yield was negatively correlated with root gloindg and leaf blight disease. In the case of root lodging, Al toxicity causes short, thickd a unnder developed roots and plants, thus 170 reducing nutrient uptake (Sasaekt ia l., 1996). Other findings indicate that Al toxicisty t he main problem because it inhibits maize root gro wrethd,ucing the water and nutrient uptake and interferes in different physiological proces osfe csrop development (Roeyt al., 1988). Foy et al. (1978) reported that low pH contributes to soliusbei lAl and make it available in the soil for plant assimilation, causing severe adgaem to non-adapted genotypes. In this case highly susceptible varieties are likely toe h laovw yield and be susceptible to lodging. Diseases affect photosynthetic area and resulot win clarbohydrate accumulation in the grain, hence the negative correlation. Correlation among the traits showed that grainu treex twas significantly and positively correlated with shelling percentage but was neeglayt ivand significantly correlated with plant vigour. Grain texture, refers to hardnesisn t(infless) or softness (dent) of the kernel. Flint kernels have higher percentage of amylope scttainrch formed by branched chain of glucose molecules of high molecular weight. Brangch rieinforces the caryopsis while soft kernels have relatively higher percentage of amey lsotsarch formed by straight chain of glucose molecules. Regular corn contains 72-76% loapmeyctin and 24-28% amylose (International Starch Institute, 2001). In terms d oisfeases, gray leaf spot was positively and significantly correlated with rust. This is oam cmon scenario where different disease pathogens tend to occur together and this call sb rfoereding for horizontal resistance through the use of inbred lines with negative GCoAr sfuch diseases. Plant vigour was positively correlated with leaf blight at optimaol ncditions but was significantly and negatively correlated with 100 seed weight. In ro twhoerds it was positively correlated with 100 seed weight as a score of 1.0 for plant vigiso ubre tter than a score of 5. The results are consistent with Welckeer t al. (2005) who reported positive correlation undeidr ascoils. Plant vigour is likely to have been affected inim ai lsar manner to grain yield under low pH soils such that hybrids that had good vigour proedu rcelatively better yields than those with low vigour. Low plant vigour under low pH reltssu in low biomass. Kochian et al. (2005) indicated symptoms of Al toxicity as reduocntsi in biomass and the number as well as the length of the roots, often combined withi nacnre ase in the mean radius and root volume; and the uptake of water and mineral nuttsri,e rnesulting in severe losses of root elongation and ultimately productivity. 171 6.7 Conclusions Low pH soil is one of the most important abiotic tfoars contributing to low maize yields in some parts of Malawi. In this study, inbred slin CeZL999601 and CML481 were identified to have higher GCA for grain yield. Inebdr lines CML312, CML448, CML172, and CML202 were also identified to have the des iGreCdA for salient traits like plant vigour, 100 seed weight, shelling percentage aanyd lgeraf spot disease respectively. Single cross hybrids CML99601/CML144, CML144/CML202, CML14/8 CML288 and CML161/CM172 were identified as good specific conmabtions for grain yield and could be used in the formation of DCs which are morel ierenst i to stress. The inbred lines and specific combinations identified in this study w bilel used in the development of maize hybrids and synthetics tolerant to low pH and dsiseesa. 6.8 References Aldrich, S.R., Scott, W.O. and E.R. Leng. 1975. Merond corn production. A & L Publication Champaign, IL Allard, R.W. and A.D. Bradshaw. 1964. Implicatioonf sg enotype-environment interactions in applied plant breeding. Crop Sci.4: 503–508. Badawy, M.E.M. 2013. Heterosis and Combining Ayb iilnit Maize using Diallel Crosses among Seven New Inbred Lines. Asian Journal of C Srcoipence 5:11-13 Baker, R.J. 1978. Issues in Diallel Analysis. CSrocpie nce 18: 533-536. Chaubey P.K. and R.P. Singh.1994. Genetic vartiay,b icliorrelation and path analysis of filled components in rice. Mandras Agr. J. 81:43780-.4 Chinnadurai, I.S. and P. Nagarajan. 2011. Intetriorenlsahip and Path-coefficient Studies for Qualitative Traits, Grain Yield and other Yield Aribttutes among MaizeZ (ea mays L.). International Journal of Plant Breeding and Gtiecns e5: 209-223 . Devi, P. and N.K. Singh., 2011. Heterosis, molerc duilvaersity, combining ability and their interrelationships in short duration maizZee a( may)s across the environments. Euphytica 178: 71-81 Falconer, D.S. and T.F.C. Mackay. 1996. Introdunc toio quantitative genetics (fourth edition). Longman Group Limited, England. 172 Foy, C. D., R. L. Chaney and M.C. White.1978. Theys piology of metal toxicity in plants. Annu. Rev. Plant Physiol. 29: 511-566. GenStat. 2013. Introduction to GenStat for Windo 1w6s..1th edition. VSN International, Hemel Hempstead, Hertfordshire HPI. IES, UK. Granados, G., S. Pandey and H. Ceballos. 1993.o Rnsees pto selection for tolerance to acid soils in a tropical maize population. Crop Scie3n3ce: 936-940. Griffing, B. 1956. Concept of general and speccifoicm bining ability in relation to diallel crossing systems. Australian Journal of BiologSiccaile nce 9: 463-49 3. Hintze, J.L. 2007. Number Cruncher Statistical eSmys t(NCSS). Kaysville, Utah. USA. International starch Institute, 2001. Maize (co rSnc)i.ence Park Aarhus, Denma rk. Iqbal, A.M., F.A. Nebvi, S.A. Wani, R. Oadil and AZ. .Dar. 2007. Combining Ability Analysis and Yield Related Traits in Maize. Intetrionnaal Journal of Plant Breeding and Genetics1: 101-10 5. Kaonga, K.K.E. and W.G. Nhlane. 2004. Evaluatio nE aorfly Maturing Maize hybrids for Low Altitude Areas of Malawi. In: D. Polland, M. Swakins, J.M. Ribaut and D. Hoisington (Eds.) Resilient Crops for Water Limi tEendvironments: Proceedings of a workshop held at Cuernavaca, Mexico. 24 – 28 2M0a0y4 , CIMMYT D.F., Mexico. pp. 104-105. Kaonga, K.K.E., D.M. Lungu and R.P. Ganunga. 20E0x7p.r essivity of Opaque 2 mutant gene in maize genotypes under low nitrogen soi lm aonidsture stressed conditions. MSc Dissertation submitted to the University of Zbaiam. Lusaka, Zambi a. Kochian, L.V., M.A., Peneros, O.A. Hoekenga. 200T5h.e physiology, genetics and molecular biology of plant aluminium resistance atonxdicity. Plant and Soil 274: 75- 195. Lerner, I M. 1954. Genetic Homeostasis. Oliver Banody d, Edinburgh. Lewontin, R C. 1956. Studies on homeostasis anedr ohzeytgosity. Amer. Nat. 90: 237– 256. Lewontin, R.C. 1957. The adaptation of populatitoon vsa rying environments. Cold spring Harb. Symp. Quant. Biol. 22: 395-408. Magnavaca, R., C.O. Gardner and R.B. Clark. 198v7a.lu Eation of inbred maize lines for aluminum tolerance in nutrient solution. Geneticp eAcsts of plant Mineral Nutrition. Martinus Nijhoff, Dordrecht, The Netherlands, pp5.5 2-265. 173 Magorokosho, C., K.V. Pixley and P. Tongoona. 2 0S0e3l.ection for drought tolerance in two tropical maize populations. African Crop Scie n1c1: 151-161. MOA (Ministry of Agriculture). 2012. Crop Estimat efosr 2012. Lilongwe. Malawi Morris, M.L., J. Risopoulos and D. Beck. 1999. Gteicn cehanges in farmer– recycled maize seed: A review of the evidence. CIMMYT EcWono. rking Paper No. 99– 107. Mexico, D.F., CIMMYT. Ndhela, T. 2012. Improvement strategies for yieoldte pntial, disease resistance and drought tolerance of Zimbabwe maize inbred line. A PhD itsh essubmitted to the University of the Free State, South Africa. Ojo, D.K., O.A. Omikunle, M.O. Ajala and S.A. Oguyboa. 2006. Heritability, character correlation and path coefficient analysis among insbixred lines of maize. World Journal of Agricultural Science 2: 352-358. Pérez, V. J.C. 2008. Adaptation of maize to aluummin itoxic soils". Part of a thesis presented to the University of São Paulo, ESALQP/ U, iSn partial fulfilment of the requirements for a doctoral degree in Agronomy. Rojas, B.A. and G.F. Sprague. 1952. A compariso vna orifance components in corn yield trials: III. General and specific combining abi liatynd their interaction with locations and years. Agron. J. 44: 462–466. Roy, A.K., A. Sharma and G. Taludker. 1988. Sompee catss of aluminium toxicity in plants. Bot. Review 54: 145-147. Sasaki, M., Y. Yamamoto and H. Matsumoto. 1996.n iLnig deposition induced by aluminium in wheat T(riticum aestivum) roots. Physiol. Plant. 96: 193–1 98. Sprague, G.F. and L.A. Tatum. 1942. General vsc.i fsicp ecombining ability in single crosses of corn. Journal of the American Socie tAy gorfonomy 34: 923–932. Sprague, G.F. and W.T. Federer. 1951. A comparoisf ovna riance components in corn yield trials II. Error, year x variety, locationv xa riety components. Agronomy Journal 43: 535-541. Srdic, J., Z. Pajic and S. Drinid-Mladenovic. 20. 0In7heritance of maize grain yield components. Maydica 52: 261–264. Statistical Analysis Software (SAS). 2013. Stactiaslt i Analysis Software (SAS) for Windows 9.0 ed., SAS, Cary, NC, USA. 174 Vacaro, E., J. Fernandes, B. Neto, D.G. Pegora.rNo., NCuss and L.H. Conceicao. 2002. Combining ability of twelve maize populations. P.e Asqgropec. Bras. Brasilia 37: 67-72. Valdivia-Bernal, R. and A.R. Hallauer. 1991. Esttiemsa of Genetic homeostasis in maize. In Rev. Brazil Genet. 14:483-500. Welcker, C. 2005. Heterosis and Combining Abifloitry Maize Adaptation to Tropical Acid Soils. Crop Science 45:2405-2413. Yan, W. and L.A. Hunt. 2002. Biplot analysis of ldleial data. Crop Science 42: 21-30. 175 CHAPTER 7 General conclusions and recommendations Maize is the number one food crop in Malawi with aannnual requirement of 2.4 million tons per year. Its production is constrained buy ma bner of abiotic factors and most of them are edaphic factors like low pH soils. This studoyo kled at the performance of maize genotypes from different genetic backgrounds foler ratonce to acid soils. The genotypes were obtained from two breeding institutions in ttrhoepical and sub-tropical regions. Two approaches were employed to evaluate their perfnocrem uander low pH soil environments. These involved the use of potassium aluminium sautlep h[KAl(SO4)2] in a nutrient solution in a glasshouse hydroponic experiment and fieladl st.r iSignificant differences were observed among the genotypes tested using KA4)l(2S sOuch that IWDC3SYNF2-B, VPO52 and LPHpop4 were found to be tolerant to ploHw. The field trial results indicated that low pH soils made a significant contributioon tthe yield and yield component reduction of 69.9%. Some genotypes that were amthoen gto p ten for field trials were also listed among the top ten in the hydroponic expenritm. Pehenotypic traits associated with grain yield, like plant vigour, 100 seed weight anudmbers of ears per plant were negatively affected by the low pH soil environmaentd can be used alongside grain yield when selecting maize genotypes which are toleroa natc itd soils. Soil analysis identified Lunyangwa Research Site soils as the most acid itch eo flow pH sites characterised. However, Tsangano Research Site was at a highiteurd ael tthan Lunyangwa (1524 masl versus 1342 masl) and was cooler than the lat8te.8r o(C2 versus 32.o0C max temperatures). In view of this, dendrogram clustering based oni reonvmental means generated by AMMI identified Tsangano as the most stressed sitee o tfw tho most acidic sites, i.e., Tsangano and Lunyanwa. As a recommendation, field trialsu sldh obe accompanied by glasshouse screening for effective selection for tolerancea ctoid soils bearing in mind the complexity of the field environment. With respect to GxE and stability analysis, geneosty pLPHpop21, VPO52, VPO72, VPO744 and VPO96 were stable across low pH andm oapl tisoil conditions. While genotype VPO97 was identified as the most unstgaebnleo type. Chitala optimal low-land site was identified as the most discriminating eronnvmi ent for the genotypes as it was 176 located at the longest distance between its maarnkedr the origin on the GGE biplot. Chitedze optimal was identified as a stable envmireont as its IPCA score and vector was near to the origin (zero). Inheritance studies identified inbred lines CZL90919 6and CML481 which had good GCA for grain yield. It was found that additive and n-aodnditive gene action was the mode of genetic inheritance for tolerance to acid soils g froarin yield and some yield related traits, recording a GCA/SCA ratio above unity while chaeraricsttics such as grain texture, stem and root lodging, gray leaf spot disease and riusseta dse recorded a GCA/SCA ratio below unity. SCA analysis identified four specific comabtiinons CML99601/CML144, CML144/CML202, CML481/ CML288 and CML161/CM172 aos ogd combinations for acid tolerance and these single-cross hybridsb wei lul sed in the development of three way and double crosses for further low pH researche pcrtso.j 177 SUMMARY In Malawi maize is grown even in marginal lands ,s otenep slopes, wet lands, rocky areas and low pH soils due to the high human populatiohnic wh excert pressure on the land. The objectives of this study were to investigate gecnaelltyi diverse maize genotypes for tolerance to low pH soil conditions. In the hydronpico experiment genotypes IWDC3SYNF2-B, VPO52, and LPHpop 4 had relativelyg hheir nett seminal root length and were considered tolerant, and DT-YSTR SYNTHE-TBI,C TZE-WPOPDTC2STR-B, TZE-YDTSTRC4-B, LPHpop3, LPHpop13, and LPHpop14 ew seernsitive or susceptible to Al toxicity. Under field conditions, genotypeLsP Hpop16, LPHpop3, VPO739, VPO5173 and LOW N POOL C3-B were identified to eblea trively tolerant to low pH soil conditions. SYNDTE–STY-W-B ranked first in terms rofot tolerant index (RTi) with a good NSRL in the glasshouse hydroponic experimnedn tt hais was followed by VPO717 which also had a relatively a better root tolera incde x and nett seminal root length. Phenotypic traits associated with grain yield, s uacsh plant vigour, seed size (100 seed weight), shelling percentage, number of ears paenr tp, lear height and plant height can be used alongside grain yield when selecting germp lfaosrm tolerance to low pH stress. In general, the effects of low pH soil conditions croibnutted to reduction in grain yields and yield components. The combined mean yield rednu cdtiuoe to low pH soil in this study was 69.9%. From AMMI and GGE analysis, genotypesH pLoPp21, VPO52, VPO72, VPO744 and VPO96 were identified as the most s.t aVbPleO097 was identified as an unstable genotype. Chitala low-land optimal sites widaentified as the most discriminating environment in terms of genotypes while Chitedzde- maltiitude optimal environment was identified as a stable environment. The diallel study revealed that additive and nodni-taivde gene actions were at play in the expression of some of the traits like grain yienludm, ber ears per plant, shelling percentage, 100 seed weight and plant vigour, while non-adde itgivene action was predominant in the inheritance of characteristics such as antheskisin-gsi linterval, plant and ear height, grain texture, stem and root lodging and gray leaf spisoet adse. Positive and highly significant GCA effects for grain yield were observed for indb rliene CZL999601 across low pH and optimal conditions. While negative and significaGnCt A effects for grain yield were 178 observed for inbred line CML161 across low pH anpdti moal conditions. SCA results indicated that single-cross hybrids CML999601/CM4L,1 4 CML144/CML202, CML481/CML288 and CML161/CM172 were best for grayiine ld across low pH and optimal conditions. At a cut-off point of 1.0 wiath c ophenetic correlation ofc o pr = 0.87, the UPGMA clustered the inbred lines based on GCA froari ng yield into two main clusters through use of Euclidean distance and standarda tdioenv i as type of scale. The pattern mostly followed the origin of the maize inbred lsin seuch that six out of seven inbred lines from CIMMYT-Colombia (tropical) were grouped in t hseecond cluster. The open pollinated inbred line varieties and specific conmabtions (single-crosses generated) identified in this study will be used in the Nataiol nMaize Breeding Programme for development of genotypes tolerant to low pH ande adsises for yield improvement and subsequent food security in the country. Key words: low pH, hydroponic, cophenetic correlation, diallel cros, spehsenotypic traits, GCA, SCA 179 OPSOMMING In Malawi word mielies selfs in marginale grond, sotpeil hellings, vleilande, klipperige gebiede en in lae pH grond verbou, weens die hvooël kbieng wat druk op die land plaas. Die doelwitte van hierdie studie was om genetiesv erdsie mielie genotipes vir verdraagsaamheid vir lae pH grondtoestande te soonedke.r In die hidroponiese eksperiment het genotipes IWDC3SYNF2-B, VPO52 en LPHpop4 rewlaeti hoër netto seminale wortel lengtes gehad en word as verdraagsaam beskou eYnS DTRT- SYNTHETIC-B, TZE- WPOPDTC2STR-B, TZE-YDTSTRC4-B, LPHpop 3, LPHpop1e3n, LPHpop14 was sensitief of vatbaar vir Al toksisiteit. Onder vteoledstande, is genotipes LPHpop16, LPHpop3, VPO739, VPO5173 en LOW N POOL C3-B ast ireefl averdraagsaam vir lae pH grondtoestande geïdentifiseer. In die glashiudirso hponiese eksperiment was SYNDTE- STY-W-B eerste in terme van wortel tolerante ind mekest 'n netto seminale wortel lengte van 2.5 cm en is gevolg deur VPO717 mʼne wt ortel tolerante indeks van 1.0 en netto seminale wortel lengte van 1.7 cm. Fenotipiese eienskappe wat geassosieer word meatn ogprbarengs, soos groeikrag, saadgrootte (100 saad gewig), saad persentas iaea, ndtiael koppe per plant, kop hoogte en plant hoogte kan gebruik word saam met graanopbsr wenagnneer kiemplasma geselekteer word vir verdraagsaamheid vir lae pH stres. In adligee meen dra die effek van lae pH grondtoestande by tot die verlaging in graanopbsrteen gen opbrengs komponente. Die gekombineerde gemiddelde opbrengs verlaging asl gg evvaon lae pH grond in hierdie studie was 69.9%. Deur die AMMI en GGE analise PisH Lpop21, VPO52, VPO72 is VPO744 en VPO96 geïdentifiseer as die mees sta gbeienloetipes. VPO097 is geïdentifiseer as 'n onstabiele genotipe. Chitala lae-ligging mopatlei omgewing is geïdentifiseer as die mees onstabiele omgewing in terme van genotipews ylt eCr hitedze middel-ligging optimale omgewing geïdentifiseer is as 'n staboiemleg ewing. Die dialleel studie het getoon dat beide additiewn en ie-additiewe geenaksʼnie r ol gespeel het by die uitdrukking van die eienskappe soosn gorpabarengs, aantal koppe per plant, saad persentasie, 100 saad gewig en groeikrag, terwey-al dndi itiewe geenaksie oorheersend was in die oorerwing van eienskappe soos antese-banatredr vai l, plant en kop hoogte, graantekstuur, stam en wortel omval en blaarvlekt es.i Positiewe en hoogs betekenisvolle 180 algemene kombineervermoë (GCA) vir graanopbren gsin isdie lae pH en optimale omgewings, vir ingeteelde lyn CZL999601 waargene temrw,yl negatief en betekenisvolle GCA effekte vir graanopbrengs waargeneem is vier tieneglde lyn CML161 vir lae pH en optimale toestande. Resultate vir spesifieke komeebrivnermoë (SCA) het aangedui dat enkelkruisbasters CML99601/CML144, CML144/CML202,M LC481/ CML288 en CML161/CM172 die beste graanopbrengs oor lae pHop etinm ale toestande gelewer het. By 'n afsny punt van 1.0 met 'n ko-fenetiese koarsriel van crop = 0.87, het die UPGMA kluster, deur die gebruik van Euklidiese afstan ds teanndaardafwyking as skaal tipe, die ingeteelde lyne gebaseer op GCA vir graanopbren gtsw ei e hoof groepe verdeel. Die groepering het die oorsprong van die ingeteelde lgyenveolg, sodanig dat ses uit sewe ingeteelde lyne van CIMMYT-Kolombië (tropiese) inie dtweede kluster groepeer. Die OPVs, ingeteelde lyne en spesifieke kombinasie lekrnukiesings wat geïdentifiseer is in hierdie studie sal gebruik word in die Nasionalee lMiei Teelprogram vir die ontwikkeling van genotipes tolerant vir lae pH en siektes en opmb rengs te verbeter sodat voedselsekuriteit in die land verseker kan word. Sleutel woord:e lae pH, hidroponiese, ko-fenetiese korrelasiael,l edei l kruise, fenotipiese eienskappe, GCA, SCA 181 APPENDICES Appendix 1R oot length measurements and derived data befodr es eavnen days after transplanting a glasshouse phoyndirco experimen t Pedigree Initial Final Control Rti % RSRLCtrl RSRLTrtd(cm) NSRL cm cm cm response (cm) cm 1 99TZEFY-STR QPM CO-B 2.8 5.0 6.7 0.7 -25.9 2.03 .191 2.2 2 DT-WSTR SYNTHETIC-B 8.7 7.4 9.1 0.8 -18.0 0.08 .1-20 0.9 3 DT-YSTR SYNTHETIC-B 2.9 1.7 4.3 0.4 -59.3 0.50 .3-90 0.6 4 EVDT-Y2000 STR QPM CO-B 6.3 7.5 8.9 0.8 -15.7 10 .4 0.19 1.2 5 EVD-W 99 STR QPM CO-B 7.2 8.5 10.8 0.8 -20.9 0 .50 0.19 1.2 6 IAR-FLINT-Q-B 2.4 2.2 3.5 0.6 -36.2 0.50 -0.04 1.0 7 IWD C3 SYN F2-B 1.4 3.5 5.1 0.7 -32.1 3.20 2.02 .0 3 8 LOW N POOL C3-B 6.1 4.3 6.1 0.7 -29.4 0.06 -0.28 0.7 9 MULTICOB EARLY DT-B 7.6 6.4 7.8 0.8 -17.4 0.04 .1-04 0.9 10 OBA SUPER1[9021-18(IITA)]-B 2.5 2.1 4.1 0.5 -54 6. 0.67 -0.07 0.9 11 OBATANPA/IWDC2SYNF2/IWDC2SY 1.4 2.9 4.0 0.7 -287 . 1.83 1.06 2.1 12 OBATANPA/IWDC2SYNF2/IWDC2SYNF2-B 3.9 2.8 4.5 0 .6 -36.2 0.19 -0.27 0.7 13 OBATANPA/TZLCOMP4C3F2/TZLC 7.0 5.8 8.5 0.7 -2 8.2 0.21 -0.16 0.8 14 OBATANPA/TZLCOMP4C3F2/TZLCOMP4C3F2-B 3.8 2.6 4.8 0.6 -43.8 0.36 -0.26 0.7 15 POP66 SR/DMR-LSRY/DMR-LSRY 2.6 2.1 3.9 0.6 -4 4.7 0.50 -0.17 0.8 16 POP66 SR/TZUTSR-WSGY/T 5.2 7.5 9.0 0.8 -16.1 4 0.7 0.47 1.5 17 SYN DTE STR-Y-B 1.7 2.4 3.5 0.7 -29.1 1.24 0.55 1.6 18 SYN DTE STY-W-B 3.4 8.5 8.2 1.1 5.4 1.39 1.49 5 2. 19 TZE COPM3 DTV C2 F2-B 1.4 1.8 2.9 0.7 -34.7 1 .02 0.27 1.3 20 TZE E-WPOP X LD(SET2)-B 1.3 2.7 3.7 0.7 -26.4 971 . 1.21 2.2 21 TZE-W POP DTC2 STR-B 4.5 2.7 5.6 0.5 -51.0 0.26 -0.39 0.6 22 TZE-WDT STR QPM-CO-B 8.5 9.4 11.5 0.8 -18.2 0 .37 0.11 1.1 23 TZE-YDT STR C4-B 5.2 3.2 4.7 0.8 -20.4 -0.07 3-80 . 0.6 24 TZE-YPOP DTC2 STR-B 1.8 2.5 3.0 0.8 -17.5 0.70 .39 0 1.4 25 VPO0721 6.1 7.6 9.2 0.8 -15.8 0.50 0.26 1.3 26 VPO5148 2.8 3.2 5.2 0.6 -37.1 1.16 0.46 1.5 182 Appendix 1 continu e 27 VPO5173 2.0 2.5 3.8 0.7 -33.2 1.37 0.57 1.6 28 VPO5179 4.8 3.5 5.8 0.6 -38.5 0.23 -0.24 0.8 29 VPO5187 3.3 4.6 5.7 0.8 -18.0 0.78 0.43 1.4 30 VPO52 1.4 3.4 4.4 0.8 -21.1 3.30 2.47 3.5 31 VPO627 1.7 3.3 4.5 0.7 -27.6 1.68 0.95 1.9 32 VPO630 1.5 2.5 3.9 0.6 -37.0 1.79 0.73 1.7 33 VPO710 5.9 4.8 7.7 0.6 -38.4 0.34 -0.17 0.8 34 VPO712 2.4 2.7 4.2 0.7 -33.8 0.77 0.16 1.2 35 VPO716 2.3 2.8 3.9 0.7 -27.9 0.79 0.27 1.3 36 VPO717 2.9 4.9 4.8 1.0 0.5 0.66 0.70 1.7 37 VPO738 1.9 3.5 4.2 0.8 -16.6 2.0 38 VPO739 5.6 4.8 7.3 0.7 -31.7 0.31 -0.12 0.9 39 VPO741 6.5 6.6 8.7 0.8 -23.3 0.36 0.03 1.0 40 VPO743 3.2 3.5 4.7 0.7 -26.1 0.65 0.21 1.2 41 VPO744 2.2 2.1 3.5 0.6 -38.7 0.66 0.02 1.0 42 VPO76 10.1 10.0 11.1 0.9 -9.0 0.11 -0.01 1.0 43 VPO86 4.2 5.4 7.0 0.8 -23.3 0.86 0.42 1.4 44 VPO96 3.4 4.8 6.6 0.7 -26.8 0.99 0.46 1.5 45 VPO97 2.4 3.3 4.4 0.8 -23.4 0.86 0.43 1.4 46 LPHpop 4 1.4 4.0 5.7 0.7 -27.8 3.31 1.99 3.0 47 LPHpop 3 5.7 3.4 6.1 0.6 -42.7 0.09 -0.41 0.6 48 LPHpop 6 1.8 3.1 4.8 0.7 -30.9 1.73 0.69 1.7 49 LPHpop 8 1.7 2.0 2.9 0.7 -30.8 0.83 0.23 1.2 50 LPHpop 9 6.0 5.6 7.1 0.8 -20.0 0.20 -0.05 0.9 51 LPHpop 10 5.3 8.1 8.0 1.0 4.0 0.50 0.56 1.6 52 LPHpop 11 8.2 7.9 10.4 0.8 -23.9 0.29 -0.02 1.0 53 LPHpop 13 4.3 2.3 4.4 0.5 -46.8 0.06 -0.44 0.6 54 LPHpop 14 6.2 4.0 6.4 0.6 -35.5 0.03 -0.36 0.6 55 LPHpop 15 2.2 2.6 3.7 0.7 -29.9 0.78 0.24 1.2 56 LPHpop 16 3.3 3.8 5.0 0.8 -21.7 0.50 0.17 1.2 57 LPHpop 17 2.8 2.7 4.2 0.7 -35.0 0.50 -0.03 1.0 183 Appendix 1 continu e 58 LPHpop 18 1.4 2.4 4.0 0.6 -38.4 2.74 1.10 2.1 59 LPHpop 19 7.4 7.2 9.4 0.8 -22.7 0.29 -0.02 1.0 60 LPHpop 20 1.9 2.9 3.5 0.8 -17.0 0.98 0.65 1.7 61 LPHpop 21 2.7 3.6 5.0 0.7 -29.0 0.91 0.38 1.4 62 LPHpop 23 1.5 1.7 3.2 0.5 -46.2 1.25 0.24 1.2 63 LPHpop 1 3.2 3.0 4.8 0.6 -37.7 0.50 -0.07 0.9 64 LPHpop 2 1.4 2.1 3.7 0.6 -41.9 1.64 0.50 1.5 65 LPHpop 7 1.8 1.5 2.7 0.5 -45.5 0.50 -0.18 0.8 66 LPHpop 12 4.4 3.9 5.6 0.7 -29.3 0.30 -0.07 0.9 67 ZM309 0.8 1.8 3.4 0.6 -38.5 3.82 1.50 2.5 68 ZM523 3.3 2.5 4.5 0.5 -46.3 0.41 -0.25 0.7 69 ZM623 3.2 3.7 5.7 0.7 -33.8 0.80 0.19 1.2 70 ZM721 3.4 4.0 6.3 0.7 -32.0 0.93 0.26 1.3 Mean 3.73 4.08 5.6 0.7 -29.1 0.89 0.32 1.3 LSD 0.95 1.2 1.87 0.22 22.01 1.27 0.88 0.88 Prob 0.001 0.001 0.001 0.001 0.001 0.001 0.001 0.001 CV (%) 15.8 18.3 20.5 19.2 46.9 88.3 171.5 41.5 Min 0.8 1.5 2.7 0.4 -59.3 -0.07 -0.44 0.6 Max 10.1 10 11.5 1.1 5.4 3.82 2.47 3.5 R-Squared 0.96 0.92 0.85 0.53 0.54 - - 0.69 Heritability 0.98 0.96 0.91 0.56 0.58 - - 0.74 ISRL = initial seminal root length, FSRL = finaml sineal Root length, Zero Al = Zero Aluminium or cronl,t RTi = root tolerance index, RSRLzero Al = rteivlae seminal root length with zero Aluminium, RSRLTRTD = relative seminaol rt olength treated with Aluminium and NSRL = net isneaml root length, LSD = Least significant diffeere, nCcV = coefficient of variation, Min = minimum, Max = maximum. 184 Appendix 2 Maize genotypes evaluated in the friieallds t2011/12 and 2012/13 Genotype number Genotype G1 DT-WSTR SYNTHETIC-B G2 EVD-W 99 STR QPM CO-B G3 IAR-FLINT-Q-B G4 IWD C3 SYN F2-B G5 MULTICOB EARLY DT –B G6 SYN DTE STY-W-B G7 OBA SUPER1(9021-18(IITA))-B G8 LPHpop 9 G9 POP66 SR/DMR -LSRY/DMR-LSRY G10 POP66 SR/TZUTSR-WSGY/T G11 LPHpop 7 G12 TZE-YDT STR C4-B G13 VPO0721 G14 TZE E -WPOP X LD(SET2)-B G15 VPO5173 G16 VPO717 G17 VPO5187 G18 VPO52 G19 VPO630 G20 LPHpop 21 G21 VPO741 G22 VPO739 G23 LPHpop 18 G24 LPHpop 16 G25 VPO743 G26 VPO744 G27 LOW N POOL C3-B G28 LPHpop 10 G29 VPO76 G30 LPHpop 3 G31 LPHpop 20 G32 LPHpop 2 G33 VPO710 G34 VPO712 G35 LPHpop 8 G36 OBATANPA/TZLCOMP4C3F2/TZLCOMP4C3F2-B G37 VPO738 G38 LPHpop 13 G39 LPHpop 15 G40 VPO86 G41 VPO96 G42 VPO97 G43 ZM309 Check 1 G44 ZM523 Check 2 G45 ZM721 Check 3 185 Appendix 3 Soil sampling data Ca(µg Lab no. Location Depth (cm) pH %OC %OM %N P(µg g-1) g-1) Mg(µg/g) % Al 97135 Tsangano 0-15 5.27 0.20 0.35 0.02 16.30 1.72 0.29 0.4 97136 " " 15-30 5.23 1.78 3.06 0.15 5.68 1.80 0.29 0.8 97137 " " 0-15 5.35 1.69 2.91 0.15 28.32 2.51 0.39 0.4 97138 " " 15-30 5.41 2.16 3.72 0.19 46.77 2.48 0.47 2.4 97139 " " 0-15 5.44 1.81 3.11 0.16 5.64 1.84 0.44 0.4 97140 " " 15-30 5.49 1.98 3.41 0.17 5.36 2.25 0.34 0.8 97141 Bvumbwe 0-15 5.65 1.49 2.56 0.13 44.32 3.22 0.98 0.6 97142 " " 15-30 5.74 1.05 1.81 0.09 62.34 2.90 0.79 0.6 97143 " " 0-15 5.67 1.46 2.51 0.13 69.11 3.03 0.81 0.6 97144 " " 15-30 5.57 1.40 2.41 0.12 41.40 2.46 0.56 0.2 97145 " " 0-15 5.68 1.17 2.01 0.10 80.30 3.88 1.06 0.6 97146 " " 15-30 5.67 1.02 1.76 0.09 83.79 3.08 0.87 0.2 97147 Lunyangwa 0-15 4.38 1.83 3.16 0.16 21.86 0.65 0.13 1.2 97148 " " 15-30 4.62 1.57 2.71 0.14 11.92 0.60 0.11 1.0 97149 " " 0-15 4.59 0.87 1.51 0.08 30.52 0.71 0.13 0.8 97150 " " 15-30 4.42 0.70 1.21 0.06 12.56 0.44 0.101 0.2 97151 " " 0-15 5.11 1.81 3.11 0.16 20.53 2.08 0.36 0.6 97152 " " 15-30 4.49 0.87 1.51 0.08 6.90 0.69 0.13 0.6 97153 Bembeke 0-15 5.10 1.98 3.41 0.17 20.77 2.00 0.41 0.8 97154 " " 15-30 4.95 0.84 1.46 0.07 7.17 0.65 0.11 0.6 97155 " " 0-15 5.15 1.66 2.86 0.14 11.83 2.09 0.43 0.4 97156 " " 15-30 5.06 1.81 3.11 0.16 20.10 1.98 0.38 1.0 97157 " " 0-15 5.14 1.63 2.81 0.14 15.53 2.05 0.46 1.0 97158 " " 15-30 5.10 1.60 2.76 0.14 18.57 2.28 0.49 1.0 186 Appendix 4 Eigenvectors for the measured and dde rdiavtea at low pH environments across two seasons AD ASI DS EH EPP LB GLS GT HC GY MSV PH RE RL RUST SH SL SWT AD 1.000 -0.045 0.941 -0.032 0.058 -0.059 -0.081 000 . 0.137 0.047 -0.123 -0.069 -0.042 0.087 0.187 0580 . 0.044 -0.010 ASI -0.045 1.000 0.295 0.008 0.035 0.050 0.091 00 .000.176 -0.091 0.095 -0.023 -0.130 -0.072 0.021 370 .0 0.044 -0.030 DS 0.941 0.295 1.000 -0.028 0.067 -0.039 -0.046 000 .0 0.191 0.014 -0.085 -0.074 -0.084 0.059 0.186 680 .0 0.057 -0.010 EH -0.032 0.008 -0.028 1.000 0.018 0.022 -0.076 000 .0 0.012 -0.026 0.058 0.680 0.042 -0.082 -0.040 160 .0 0.004 0.182 EPP 0.058 0.035 0.067 0.018 1.000 0.016 -0.125 00 .0-00.149 0.653 -0.082 0.061 -0.068 -0.132 -0.073 9930 . -0.030 -0.130 LB -0.059 0.050 -0.039 0.022 0.016 1.000 0.380 0 .00-0.063 0.059 0.229 0.008 -0.095 0.025 0.031 0 .01-30.080 -0.030 GLS -0.081 0.091 -0.046 -0.076 -0.125 0.380 1.000. 0000 -0.031 -0.048 0.249 -0.119 -0.036 0.100 0.040 -0.124 0.038 -0.220 GT 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000.0 00 0.000 0.000 0.000 0.000 0.000 0.000 0.000 000 .0 0.000 HC 0.137 0.176 0.191 0.012 -0.149 -0.063 -0.031 000 .0 1.000 -0.225 -0.113 0.039 0.066 0.135 0.176 4-40 .1 0.052 0.094 GY 0.047 -0.091 0.014 -0.026 0.653 0.059 -0.048 0 0 .0 -0.225 1.000 -0.091 0.007 -0.113 -0.103 -0.101 .6430 0.007 -0.210 MSV -0.123 0.095 -0.085 0.058 -0.082 0.229 0.249 000 . -0.113 -0.091 1.000 -0.071 -0.084 0.044 -0.132 -0.087 -0.140 0.048 PH -0.069 -0.023 -0.074 0.680 0.061 0.008 -0.1190 000 . 0.039 0.007 -0.071 1.000 0.040 -0.010 -0.127 0620 . -0.020 0.163 RE -0.042 -0.130 -0.084 0.042 -0.068 -0.095 -0.0306.0 00 0.066 -0.113 -0.084 0.040 1.000 -0.001 0.033 -0.067 -0.000 0.096 RL 0.087 -0.072 0.059 -0.082 -0.132 0.025 0.100 0 0 .0 0.135 -0.103 0.044 -0.010 -0.001 1.000 0.181 14-30 . -0.020 0.061 RUST 0.187 0.021 0.186 -0.040 -0.073 0.031 0.0400 000 . 0.176 -0.101 -0.132 -0.127 0.033 0.181 1.000 .07-00 0.142 0.175 SH 0.058 0.037 0.068 0.016 0.993 0.013 -0.124 0 .0-000.144 0.643 -0.087 0.062 -0.067 -0.143 -0.070 001 .0 -0.030 -0.120 SL 0.044 0.044 0.057 0.004 -0.035 -0.085 0.038 00 .000.052 0.007 -0.135 -0.016 -0.003 -0.023 0.142 02-70 . 1.000 0.123 SWT -0.006 -0.026 -0.014 0.182 -0.132 -0.034 -0 .2107.000 0.094 -0.211 0.048 0.163 0.096 0.061 0.175 0.11-9 0.123 1.000 AD = days to anthesis (days), ASI = anthesis-sgi liknitnerval (days), DS = days to silking (days),= E eHa r height (cm), EPP = ears per plant (#), LBe a=f lblight disease (1-5), GLS = gray leaf spot disease (1-5), GT = grain textu-r5e) ,( 1HC = husk cover, GY = grain yield (kg- 1h),a MSV = maize streak virus disease (1-5), PH =n pt lhaeight (cm), RE = rotten ears (#), RL = root lodging (#), Rust = rust diseas5e) ,( 1S-H = shelling percentage, SL = stem lodgin, gS W(#T) = 100 seed weight (g), # = number. 187 Appendix 5 Eigenvectors from the principal compto anneanlysis for grain yield and agronomic at opt eimnvailronments across two seasons GY AD DS EH EPP LB GLS GT HC MSV PH RL RE RUST SH SL SWT VIG AD -0.06 1.00 0.98 0.13 0.09 -0.03 0.00 -0.04 0.06 -0.07 0.07 -0.04 -0.05 -0.06 0.01 -0.03 -0.09 0.11 DS -0.04 0.98 1.00 0.14 0.08 -0.04 0.02 -0.06 0.08 -0.07 0.06 -0.04 -0.03 -0.05 0.01 -0.01 -0.09 0.28 EH -0.05 0.13 0.14 1.00 0.04 -0.07 -0.03 0.05 - 0.15 0.06 0.63 -0.02 0.10 -0.07 -0.04 -0.04 0.12 0.21 EPP -0.08 0.09 0.08 0.04 1.00 0.14 0.06 -0.31 -0.01 0.09 0.08 0.04 -0.17 -0.03 -0.04 0.09 0.02 0.00 LB -0.09 -0.03 -0.04 -0.07 0.14 1.00 0.21 0.03 2-0 .0 -0.15 -0.01 -0.04 0.05 0.17 0.02 -0.05 -0.07 1- 0.5 GLS 0.01 0.00 0.02 -0.03 0.06 0.21 1.00 -0.02 - 0.04 -0.11 0.15 0.06 -0.02 -0.09 0.01 -0.01 -0.06 0.00 GT 0.04 -0.04 -0.06 0.05 -0.31 0.03 -0.02 1.00 4- 0.0 -0.12 0.05 0.00 0.01 -0.07 0.01 -0.12 -0.10 0.00 HC 0.10 0.06 0.08 -0.15 -0.01 -0.02 -0.04 -0.04 0 1.0 -0.16 -0.19 0.17 0.11 0.06 0.01 0.00 -0.06 0.00 GY 1.00 -0.06 -0.04 -0.05 -0.08 -0.09 0.01 0.04 0 0.1 0.12 -0.09 -0.01 0.03 0.09 0.10 -0.05 0.25 0.11 MSV 0.12 -0.07 -0.07 0.06 0.09 -0.15 -0.11 -0.12 .16-0 1.00 0.05 -0.04 0.00 -0.16 -0.05 0.01 0.14 0.00 PH -0.09 0.07 0.06 0.63 0.08 -0.01 0.15 0.05 -0.19 0.05 1.00 -0.02 -0.07 -0.20 -0.05 -0.03 0.13 0.00 RL -0.01 -0.04 -0.04 -0.02 0.04 -0.04 0.06 0.00 7 0.1 -0.04 -0.02 1.00 -0.04 -0.03 -0.03 -0.07 -0.10 13 0. RE 0.03 -0.05 -0.03 0.10 -0.17 0.05 -0.02 0.01 0.11 0.00 -0.07 -0.04 1.00 0.07 0.03 -0.01 -0.05 0.00 RUST 0.09 -0.06 -0.05 -0.07 -0.03 0.17 -0.09 -0.07 0.06 -0.16 -0.20 -0.03 0.07 1.00 0.01 -0.12 0.21 69 0. SH 0.10 0.01 0.01 -0.04 -0.04 0.02 0.01 0.01 0.01 0.05- -0.05 -0.03 0.03 0.01 1.00 0.11 -0.04 -0.09 SL -0.05 -0.03 -0.01 -0.04 0.09 -0.05 -0.01 -0.12 .000 0.01 -0.03 -0.07 -0.01 -0.12 0.11 1.00 0.17 32- 0. SWT 0.25 -0.09 -0.09 0.12 0.02 -0.07 -0.06 -0.10 .06- 0 0.14 0.13 -0.10 -0.05 0.21 -0.04 0.17 1.00 0.00 VIG 0.11 0.11 0.28 0.21 0.00 -0.51 0.00 0.00 0.00 .00 0 0.00 0.13 0.00 0.69 -0.09 -0.32 0.00 1.00 GY = grain yield (kg h-a1), AD = days to anthesis (days), DS = days ton sgi l(kdiays), EH = ear height (cm), EPP = ears pern tp (la#), LB = leaf blight disease (1-5), GLS = glreaayf spot disease (1-5), GT = grain texture (1-5), HhCu =s k cover, MSV = maize streak virus (1-5), PH a=n pt lheight (cm), RL = root lodging (#), RE = rot eteanrs (#), Rust = rust disease (1-5), SH = shelling percentage, SL = stem lod(g#i)n, gS WT = 100 seed weight (g), VIG = vigor, # =m nbuer. 188 Appendix 6 Soil analytical data interpretation ge uid Threshold values i) Phosphorus ug-1 g (Mehlich3) rating <8 very low 9 - 18 low 19 - 25 medium (adequate range) 25 - 33 high (adequate range) >34 very high ii) Potassium cmol k-1g(Mehlich3) rating <0.05 very low 0.06 - 0.10 low 0.11 - 0.04 medium (adequate range) 0.50 - 0.80 high >1.00 very high iii) Zinc ug g-1 (Mehlich3) rating <1.00 very low 1.00 - 1.50 low 1.60 - 2.50 medium (adequate range) 2.5 - 3.00 high (adequate range) >2.5 very high iv) Boron ug -g1 (Mehlich3) rating <0.70 low 0.8 - 1.40 medium 1.40 - 2.50 high >2.5 very high v) Copper ug -g1 (Mehlich3) rating <0.30 low 0.4 - 0.8 medium 0.9 - 2.50 high >2.5 very high vi) Manganese cmol k-1g (Mehlich3) rating <0.20 very low 0.2 - 0.5 low 0.6 - 3.9 high >4.0 very high 189 Appendix 6 continued NB: Critical levels for Mn is 3.0 ug- 1g and Ca is 2.0 cmol k-1g using Mehlich3 (Melich 1984) vii) Soil pH In water In CaC2l Rating <4.5 <4.0 very strongly acid 4.5 - 5.0 4.0 - 4.45 strongly acid 5.1 - 5.5 4.5 - 4.95 acid 5.6 - 6.05 0 - 5.45 moderately acid 6.1 - 6.5 5.5 - 5.95 slightly acid 6.6 - 7.0 6.0 - 6.45 almost neutral 7.1 -7.5 6.5 - 6.95 very slightly alkaline 7.6 - 8.0 7.0 - 7.45 slightly alkaline >8.0 -7.45 - 7.95 alkaline / moderately alka line >8.5 >8.00 strongly alkaline vii) Total Nitrogen% Rating <0.08 very low 0.08 - 0.12 low 0.12 - 0.2 medium 0.20 - 0.30 high >0.3 very high vii) % Carbon organic matter % Rating <0.88 1.5 low 0.88 - 2.35 1.5 - 4.0 medium >2.35 >4.0 high Notes: Soil reaction (pH) is determined in wateor.n N acidifying fertilisers such as CAN are recommended below pH 5.5 in water (4.5 in C2)a. CDlolomitic lime application (1-2 t ha-1). Farm yard manure is recommended at 5 tons per t hea sifoil test measures less than 1.0% OM (0.58% OC) and 2.5 tons per ha when thle t essoti is between 1.0 and 1.5% OM (0.58-0.87% OC). 190 Appendix 7 Mean performance for grain yield acrfoussr optimal environments combined for 2011/122 a0n1d2 /13 seasons Genotypes GY AD ASI DS EH EPP LB HC GLS GT MSV PH ER RL Rust SL SWT SH% VIG G1 3100 66.1 2.9 67.8 75 1.0 1.8 1.2 1.7 1.6 1.1 1 170.6 0.1 1.7 0.2 30.8 77.3 2.3 G2 2326 62.1 1.2 64.5 69 1.0 2.0 1.3 1.9 1.8 1.3 6 151.0 0.5 1.5 1.5 29.3 67.4 2.1 G3 3014 64.1 0.8 65.7 75 1.2 1.4 1.3 1.7 1.6 1.3 3 160.7 0.5 1.5 1.0 30.1 76.4 2.0 G4 3311 65.0 1.3 66.9 73 1.1 1.9 1.2 1.3 1.8 1.1 5 160.6 0.3 1.7 0.2 28.3 80.6 2.0 G5 3921 68.3 1.0 70.0 87 1.1 1.9 1.9 1.5 1.9 1.1 9 180.7 0.8 1.4 1.0 32.1 79.6 1.9 G6 3505 67.2 0.8 69.0 85 1.2 2.0 1.4 1.9 2.0 1.1 5 180.8 0.9 1.6 0.6 31.2 79.6 1.9 G7 2663 65.6 1.4 67.6 78 1.1 1.9 0.9 1.4 2.0 1.3 1 170.8 0.7 1.5 0.6 30.9 80.0 2.0 G8 3965 65.9 1.1 67.6 85 1.2 1.9 1.6 1.9 1.7 1.2 2 180.7 1.4 2.0 1.2 30.1 80.9 1.9 G9 3254 66.0 0.9 67.4 80 1.1 1.9 1.2 1.5 2.0 1.3 5 170.9 0.6 1.6 0.7 29.3 75.7 1.9 G10 3817 67.7 2.0 70.2 86 1.1 2.0 0.9 1.9 2.0 1.2 83 1 0.8 0.6 1.7 0.2 31.1 81.8 2.0 G11 2517 67.8 1.0 69.4 76 1.1 2.0 1.3 1.8 1.8 1.4 69 1 0.7 0.6 1.6 0.5 26.7 77.5 2.1 G12 3687 66.3 1.1 68.0 78 1.1 1.9 1.1 1.8 1.8 1.1 73 1 0.7 0.4 1.9 0.0 26.1 78.3 2.0 G13 3444 66.5 0.8 68.1 82 1.2 1.6 1.4 2.1 1.8 1.2 73 1 1.4 0.8 1.3 0.2 28.6 80.8 2.2 G14 3189 65.5 1.7 67.9 82 1.0 2.0 1.1 1.8 1.8 1.0 77 1 1.1 0.9 1.6 0.2 30.2 79.4 2.0 G15 3319 63.6 1.3 65.6 73 1.1 2.0 1.7 1.7 1.8 1.1 65 1 0.9 1.2 1.8 0.9 29.5 79.0 2.3 G16 2901 63.2 0.8 64.9 80 1.0 1.8 2.1 1.8 1.8 1.1 73 1 0.8 0.5 1.8 0.6 32.3 77.7 2.0 G17 3143 65.3 1.3 67.3 75 1.2 1.8 1.5 1.3 1.6 1.1 64 1 0.5 0.3 1.4 0.8 26.2 83.4 2.4 G18 3633 63.3 1.2 65.3 79 1.2 1.4 1.1 1.5 1.7 1.1 73 1 1.1 0.9 1.5 1.0 32.1 80.8 2.1 G19 3179 64.9 1.0 66.8 79 1.1 1.9 2.0 1.8 2.0 1.1 67 1 0.8 1.5 2.1 0.6 29.4 86.5 2.1 G20 4489 65.0 1.0 66.7 81 1.1 1.9 1.3 1.5 1.8 1.2 73 1 1.1 1.4 1.3 0.4 32.0 83.0 1.5 G21 3307 65.1 1.0 66.7 83 1.2 2.4 1.5 1.6 1.8 1.0 76 1 0.7 1.1 1.7 1.8 30.9 80.3 2.0 G22 3573 62.7 1.4 64.7 75 1.1 1.9 1.2 1.5 1.8 1.3 70 1 0.8 0.2 1.7 0.7 30.3 80.1 2.5 G23 3160 61.6 1.2 63.4 70 1.1 2.1 1.1 1.9 2.0 1.0 63 1 0.7 0.6 2.0 0.6 28.6 79.0 2.0 G24 2983 59.1 1.3 61.1 72 1.1 2.2 1.8 1.3 2.0 1.1 67 1 0.7 0.7 1.7 0.8 29.1 75.9 2.2 G25 3396 63.3 0.9 65.2 82 1.1 1.7 1.7 1.7 2.1 1.4 77 1 1.5 0.7 1.8 0.7 30.7 81.0 1.7 G26 3611 64.4 1.2 66.2 75 1.1 2.1 1.3 1.6 2.0 1.1 74 1 1.3 0.8 1.5 0.2 30.5 77.4 1.9 G27 3965 66.4 1.4 68.4 82 1.2 1.9 1.6 1.7 2.1 1.2 79 1 0.8 0.3 1.6 0.1 30.7 79.8 1.9 G28 3727 62.0 1.0 63.8 75 1.1 2.0 1.3 1.8 2.1 1.2 71 1 0.5 1.0 1.7 0.5 27.4 82.6 1.9 G29 3384 66.8 1.3 68.3 78 1.1 2.0 1.6 1.5 2.0 1.4 76 1 0.9 0.6 1.6 0.5 29.7 79.3 2.1 191 Appendix 7 continue d G30 3693 65.7 0.9 67.3 82 1.1 1.9 1.5 1.8 1.8 1.2 79 1 0.8 0.4 1.6 0.4 30.9 79.2 2.0 G31 3820 64.8 1.0 66.6 82 1.1 2.2 1.3 1.8 1.8 1.2 76 1 0.8 0.3 1.7 1.0 27.6 80.4 2.0 G32 3744 62.2 1.0 64.0 80 1.1 2.0 1.8 1.5 1.9 1.4 78 1 0.5 0.6 1.6 0.6 28.6 85.4 2.0 G33 4003 61.3 1.2 63.4 75 1.2 1.8 1.5 1.4 1.9 1.6 68 1 0.7 1.0 1.7 0.9 29.5 82.1 1.9 G34 3223 60.7 0.9 62.5 78 1.2 2.2 1.3 1.6 2.7 1.0 74 1 0.7 1.8 1.9 0.3 25.9 80.2 2.1 G35 3563 64.2 1.0 65.9 81 1.1 1.8 2.6 1.6 2.0 1.1 78 1 0.7 0.7 1.6 1.2 29.9 77.8 2.0 G36 3757 62.9 1.1 64.9 83 1.1 1.9 1.3 1.6 2.0 1.3 79 1 0.8 0.8 1.9 0.7 28.6 79.5 2.0 G37 2594 65.3 1.3 67.2 66 1.1 1.7 1.0 1.5 2.1 1.2 62 1 0.6 0.2 1.9 0.9 30.4 74.1 2.0 G38 2413 69.0 1.1 70.3 78 1.0 1.9 1.0 1.4 1.9 1.4 75 1 0.5 0.8 1.4 0.8 30.5 72.8 2.0 G39 3902 64.1 0.6 65.6 70 1.1 2.0 1.2 1.5 1.9 1.1 61 1 0.9 0.6 1.8 0.8 31.4 79.3 2.1 G40 2295 66.5 1.1 68.0 73 1.1 1.9 1.2 1.4 1.9 1.2 69 1 0.8 0.6 1.4 0.5 33.2 77.8 2.0 G41 3130 60.4 0.6 62.0 70 1.0 1.5 1.3 1.4 1.5 1.0 64 1 0.9 0.2 1.9 0.3 27.5 82.2 2.3 G42 2996 63.2 1.0 64.9 71 1.0 1.8 0.9 1.6 1.7 1.2 68 1 0.5 1.6 1.7 0.5 31.3 78.7 2.2 G43 3750 61.9 0.9 63.8 69 1.1 2.0 1.3 1.6 1.9 1.2 62 1 0.9 0.2 1.7 1.7 31.0 77.4 1.9 G44 3685 64.3 1.2 66.5 75 1.0 2.0 1.4 1.8 1.7 1.2 72 1 0.8 0.5 1.5 1.0 32.1 79.4 2.0 G45 3735 66.2 0.8 68.0 89 1.1 1.8 1.4 1.3 2.5 1.7 86 1 0.7 0.2 1.8 0.9 33.7 76.7 1.8 Mean 3373 64.5 1.1 66.3 77.6 1.1 1.9 1.4 1.6 1.9 2 1.172.3 0.8 0.7 1.7 0.7 29.9 79.2 2.0 LSD 608 1.7 1.3 1.7 8 0.1 0.3 0.8 0.3 0.5 0.3 12 4 0. 1.2 0.3 0.7 2.9 5.9 0.3 CV 31.8 4.6 194 4.6 18.2 15 27.1 103 36 40 35.9 1 12. 80 302 31.3 169 16.8 13.1 27.3 SE 1072 3.0 2.2 3.1 14.1 0.2 0.5 1.4 0.6 0.9 0.4 .8 20 0.7 2.1 0.5 1.2 5.1 10.4 0.6 Min 2295 59 1 61 66 1 1 1 1 1 1 156 1 0 1 0 26 67 2 Max 4489 69 3 70 89 1 2 3 2 3 2 189 2 2 2 2 34 86 2 LSD = Least significant differen cCeV, = coefficient of variatio nS,E = error , Min = minimum, Max =m aximum ,GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = anthesis-silking interval (days), DS = daysl ktoin sgi (days), EH = ear height (cm), EPP = ears plearn t (#), LB = leaf blight disease (1-5), HC = kh ucosver, GLS = gray leaf spot disease (1-5), GT = grain texture (1-5), MSV = mea siztreak virus disease (1-5), PH = plant height) ,( RcmE = rotten ears (#), RL = root lodging (#), Rt =u srust disease (1-5), SL = stem lodging (#), SWT = 100 seed weight (g), ShHe =ll isng percentage, VIG = vigour, # = number. 192 Appendix 8 Mean performance for grain yield ande ro tahgronomic traits across four low pH environm ceonmtsbined for 2011/12 and 2012/13 seaso ns Genotypes GY AD ASI DS EH EPP LB HC GLS GT MSV PH ER RL Rust SL SWT SH% VIG G1 1417 84.1 2.8 87.0 52.0 0.9 2.1 1.3 2.4 2.4 1.2 136 1.4 1.7 1.2 1.2 23 67 2.8 G2 1282 82.4 2.3 84.8 48.2 0.8 2.4 1.6 1.8 2.4 1.2 127 0.9 2.5 1.4 1.7 22 65 3.0 G3 1562 81.4 2.5 84.0 49.4 0.8 2.1 1.7 2.0 2.5 1.2 132 0.8 1.9 1.1 1.7 26 75 2.6 G4 1372 83.5 3.4 87.3 48.6 0.8 2.3 2.2 2.2 2.7 1.1 127 1.7 1.8 1.4 0.9 23 68 2.4 G5 1517 85.9 2.0 87.9 44.3 0.8 2.3 2.1 1.9 2.4 1.2 129 1.2 1.8 1.3 1.5 24 71 2.2 G6 1655 82.0 2.0 83.1 44.4 0.9 2.4 1.3 2.7 2.8 1.2 129 1.6 1.4 1.2 0.7 22 68 2.7 G7 1527 80.4 3.3 83.0 46.2 0.8 2.3 1.2 2.1 2.7 1.2 301 1.4 1.7 1.1 1.1 24 73 2.9 G8 1488 83.3 2.5 85.8 45.4 0.8 2.7 1.1 2.0 2.6 1.2 128 0.9 1.4 1.0 1.8 23 65 2.6 G9 1537 84.6 2.3 86.9 46.9 0.9 2.2 0.8 1.9 2.5 1.2 132 1.0 1.4 1.4 1.0 22 66 2.9 G10 1314 84.6 2.8 87.5 46.2 1.0 2.3 1.4 2.1 2.9 1.3123 0.9 2.3 1.4 1.3 19 68 2.5 G11 1502 82.7 2.1 84.8 38.0 0.8 2.1 1.5 2.3 2.5 1.2120 0.9 1.2 1.3 1.5 20 72 2.8 G12 1417 82.5 2.7 85.0 43.8 0.9 2.4 1.3 2.7 2.9 1.3129 1.3 1.7 1.5 1.0 20 68 3.1 G13 1489 82.0 3.5 85.8 50.1 0.9 2.4 1.9 2.1 2.8 1.3127 1.4 1.4 1.2 2.2 22 69 3.0 G14 1362 83.6 2.7 86.3 46.5 0.8 2.2 2.5 2.3 2.3 1.2148 0.9 2.0 1.0 1.3 22 68 2.7 G15 1746 82.5 3.5 86.0 48.6 0.9 2.3 1.7 2.4 2.7 1.1132 1.2 2.2 1.5 1.4 22 70 2.8 G16 1650 80.8 3.1 82.9 38.9 0.8 1.9 1.9 2.1 2.6 1.3116 1.0 1.3 1.4 1.6 24 71 2.7 G17 1434 82.9 3.4 86.3 46.0 0.9 2.4 1.9 2.3 2.7 1.2124 1.5 2.0 1.2 1.5 23 71 2.9 G18 1539 83.3 2.0 85.3 50.4 0.8 2.2 1.6 1.9 2.5 1.2128 0.9 1.6 1.2 2.2 22 70 2.6 G19 1501 82.9 1.7 84.3 48.8 0.8 2.2 1.6 2.2 2.8 1.2127 1.2 2.0 1.4 1.4 21 74 3.2 G20 1466 84.2 2.6 87.1 45.7 0.8 1.8 1.9 2.3 2.6 1.2122 0.9 1.7 1.2 1.3 24 72 2.4 G21 1602 83.8 2.8 86.6 42.6 0.9 2.3 1.7 2.0 2.9 1.1126 1.0 1.8 1.4 1.0 22 71 2.6 G22 1753 81.6 2.5 84.8 48.2 0.9 2.2 2.1 2.1 2.4 1.2126 1.2 1.7 1.2 1.1 23 70 2.7 G23 1496 82.6 2.1 85.4 43.8 0.9 2.2 1.8 2.2 2.3 1.3122 1.6 2.2 1.2 1.2 24 71 3.0 G24 1792 79.7 3.0 83.0 51.2 0.8 2.4 1.6 2.3 2.4 1.1134 1.4 2.3 1.5 1.2 24 73 2.7 G25 1401 84.5 2.7 86.4 50.8 0.9 2.2 1.5 1.9 2.5 1.2141 1.1 1.5 1.3 1.1 23 69 2.5 193 Appendix 8 continued G26 1575 84.5 3.2 87.7 47.5 0.8 2.6 2.2 2.1 1.9 1.3 125 1.2 1.3 1.2 0.9 23 68 2.7 G27 1701 84.7 2.0 86.7 49.8 0.9 2.1 1.0 1.9 2.8 1.2 129 1.2 1.3 1.1 1.7 27 69 2.4 G28 1513 84.5 2.3 86.8 47.9 0.8 2.1 1.7 2.1 3.0 1.2 127 0.9 2.0 1.4 1.1 24 67 2.7 G29 1491 85.2 2.4 87.4 52.4 0.9 2.2 2.4 1.8 2.4 1.5 139 1.0 1.3 1.1 1.4 22 70 3.2 G30 1765 82.5 2.9 85.4 44.0 0.9 1.9 2.4 2.1 2.3 1.3 119 1.1 2.0 1.3 0.6 24 74 2.1 G31 1195 80.3 2.6 82.9 45.4 0.9 2.0 2.3 2.1 2.6 1.3 123 1.3 2.2 1.3 0.9 22 64 3.1 G32 1448 80.7 2.7 83.8 51.6 0.8 2.1 1.0 2.1 2.7 1.4 132 1.3 1.8 1.2 1.1 23 68 2.7 G33 1385 83.0 3.0 86.5 46.6 0.8 2.0 1.2 2.1 2.5 1.2 127 1.0 2.4 1.3 1.3 22 70 2.5 G34 1599 81.6 3.0 84.0 44.7 0.8 2.0 1.1 1.8 1.8 1.3 123 0.9 1.7 1.3 1.1 22 70 2.7 G35 1396 86.3 3.0 89.3 45.2 0.8 2.1 2.3 2.3 3.2 1.1 126 1.4 2.0 1.2 1.1 26 69 2.8 G36 1432 83.6 3.2 86.4 46.0 0.7 2.3 1.9 2.3 2.3 1.3 123 1.5 2.4 1.5 0.9 21 70 2.6 G37 1409 82.5 2.2 85.1 47.1 0.8 1.9 2.6 2.1 2.8 1.2 129 1.1 1.3 1.3 1.4 20 71 2.7 G38 1436 85.8 1.6 87.4 46.8 0.8 2.2 1.9 2.1 2.2 1.3 121 1.5 1.5 1.1 1.1 22 71 2.6 G39 1429 82.5 3.4 86.1 49.6 0.8 2.0 1.3 2.2 2.4 1.3 130 1.2 2.7 1.2 1.7 22 68 2.8 G40 1475 84.8 2.3 87.3 47.4 0.8 2.2 2.1 1.8 2.3 1.2 130 0.9 2.3 1.3 0.7 21 69 2.5 G41 1376 80.3 2.8 83.5 40.6 0.8 2.0 1.9 3.7 2.6 1.3 118 1.0 1.8 1.2 1.0 21 67 2.8 G42 1406 82.7 2.8 85.9 53.9 0.8 2.3 1.8 1.6 2.5 1.2 139 1.5 1.8 1.2 1.2 22 68 3.2 G43 1418 84.6 3.1 89.0 39.5 0.9 2.1 2.6 2.0 2.0 1.4 121 1.3 1.6 1.2 1.0 22 66 3.2 G44 1359 83.9 2.4 86.7 48.1 0.9 2.1 1.2 2.1 2.4 1.3 135 0.9 1.8 1.4 1.0 23 67 3.3 G45 1447 83.8 2.9 86.7 45.4 0.8 1.9 1.5 1.9 2.4 1.2 133 1.0 1.3 1.1 0.7 25 66 2.5 Mean 1491 83.1 2.7 85.8 46.8 0.9 2.2 1.7 2.1 2.5 2 1. 131.9 1.2 1.8 1.3 1.2 22.6 69.3 2.7 LSD 358 2.4 1 2.4 7.7 0.3 0.4 0.8 0.7 0.4 0.3 58 7 0. 0.9 0.3 0.9 3 8 0.4 CV 42.4 5.1 67 5 28 54 35 83 59 28 37 77 103 92.14 .6 4 127.6 20.3 19 27.4 SE 631 4.2 1.8 4.3 13.6 0.5 0.8 1.4 1.3 0.7 0.5 .51 01 1.2 1.7 0.6 1.6 4.6 13 0.8 Min 1195.0 79.7 1.6 82.9 38.0 0.7 1.8 0.8 1.6 1.8 .1 1 116.0 0.8 1.2 1.0 0.6 19.0 64.0 2.1 Max 1792.0 86.3 3.5 89.3 53.9 1.0 2.7 2.6 3.7 3.2 .5 1 301.0 1.7 2.7 1.5 2.2 27.0 75.0 3.3 LSD = Least significant difference, CV = coeffitc oief nvariation, SE = error, Min = minimum, Max = mxiamum, GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = anthesis-silking interval (days), DS = days toin sgi l(kdays), EH = ear height (cm), EPP = ears pearn pt l(#), LB = leaf blight disease (1-5), HC = hucoskv er, GLS = gray leaf spot disease (1-5), GT = grain texture (1-5), MSV = mea siztreak virus disease (1-5), PH = plant height) ,( RcmE = rotten ears (#), RL = root lodging (#), Rt =u srust disease (1-5), SL = stem lodging (#), SWT = 100 seed weight (g), ShHe =ll isng percentage, VIG = vigour, # = num.b er 194 Appendix 9 Mean performance for grain yield ando angormic traits for low N environment 2012/13 season Genotypes GY Rank AD ASI DS EH EPP LB GLS GT HC PH RL RUST SL SWT SH% G1 826 12 68.7 2.3 71.0 33.6 0.8 3.5 1.0 2.3 0.3 .3 77 6.3 1.0 3.3 32.0 64.5 G2 614 27 65.3 2.0 67.3 19.3 0.7 4.0 1.7 2.2 0.3 .7 60 8.3 1.0 0.3 25.7 59.0 G3 467 37 66.3 3.0 67.0 38.3 1.1 3.0 2.0 2.3 1.7 .0 73 5.0 1.2 1.0 27.0 60.0 G4 465 38 67.0 1.0 69.7 32.0 0.8 3.7 1.5 2.2 0.7 .7 77 6.0 1.0 0.3 29.7 50.0 G5 396 42 72.3 1.7 75.0 38.6 0.6 2.2 1.2 2.0 0.0 .7 72 3.3 1.0 1.7 29.0 51.0 G6 493 33 68.3 1.7 74.0 31.0 0.6 3.0 1.5 2.2 0.7 .3 76 1.7 1.0 1.0 29.7 43.0 G7 721 22 68.0 3.0 70.0 37.6 0.7 3.3 1.8 2.0 0.0 .3 74 6.0 1.0 1.7 30.7 56.7 G8 504 31 68.0 2.3 71.0 37.6 0.8 2.8 1.0 1.5 0.0 .7 80 1.0 1.0 0.3 35.4 52.2 G9 1146 1 68.0 2.0 70.3 56.6 0.9 3.2 1.3 1.7 1.3 8.31 0 2.3 1.0 2.0 30.0 74.5 G10 1145 2 72.3 1.3 73.0 37.3 0.8 2.8 1.5 1.7 0.0 2.3 7 4.3 1.0 1.0 34.4 70.8 G11 744 21 69.7 2.7 71.0 43.0 0.8 3.3 1.7 2.0 0.0 8.3 8 5.0 1.0 2.0 25.7 68.9 G12 471 34 67.7 2.3 70.3 44.0 0.8 3.3 1.8 2.3 0.0 7.3 7 7.3 1.0 1.7 24.7 57.2 G13 1113 3 69.0 1.3 71.3 35.6 0.7 3.3 1.7 2.2 0.7 2.3 8 4.7 1.0 0.7 28.7 70.3 G14 640 25 66.3 1.7 67.7 34.3 0.7 2.7 1.3 2.5 0.0 8.0 7 3.0 1.0 1.0 28.0 63.9 G15 377 43 66.0 2.3 67.7 32.6 0.5 3.5 1.3 2.0 0.0 9.3 6 2.0 1.0 1.7 27.7 62.4 G16 788 18 65.7 3.3 68.0 31.0 0.7 3.0 1.7 1.5 2.3 2.7 6 6.0 1.0 2.3 30.0 59.2 G17 820 14 66.3 0.7 69.7 43.0 0.8 3.0 1.8 2.2 0.0 6.0 8 1.7 1.0 0.0 28.4 69.4 G18 432 39 65.7 2.7 66.3 22.3 0.6 3.3 1.2 1.7 0.7 9.7 5 6.0 1.0 2.0 30.4 60.8 G19 974 6 67.7 2.0 70.3 41.3 1.0 4.0 1.2 2.0 2.3 .3 81 2.7 1.0 2.3 31.4 63.9 G20 959 7 65.7 1.3 67.7 45.3 0.7 3.0 1.8 2.7 1.0 .0 92 6.3 1.0 1.3 30.0 56.7 G21 938 8 66.0 1.0 67.3 33.3 0.7 3.2 2.2 1.5 0.3 .0 70 4.7 1.0 2.3 28.7 68.3 G22 678 24 63.3 1.3 64.3 49.6 0.8 3.7 1.3 1.5 0.7 00.01 3.7 1.0 4.0 28.0 63.3 G23 324 44 62.0 1.3 63.3 35.6 0.7 3.5 1.2 2.0 0.7 9.7 6 7.3 1.0 4.0 25.7 55.6 G24 1008 4 62.3 1.0 63.7 36.6 0.8 2.5 1.7 2.5 0.0 5.0 6 1.7 1.0 0.0 28.7 67.3 G25 790 16 64.0 2.7 65.0 32.3 1.0 2.8 1.3 2.2 0.0 7.7 6 2.7 1.0 1.3 34.4 65.5 195 Appendix 9 continued G26 502 32 68.3 1.3 71.0 35.3 0.6 2.7 1.3 2.5 0.0 0.7 8 1.3 1.0 2.3 30.0 57.2 G27 924 10 68.0 1.3 69.3 43.3 0.8 3.3 1.5 2.3 1.3 8.7 8 4.3 1.0 0.3 28.4 67.9 G28 799 15 66.0 2.3 67.3 43.3 0.9 3.3 1.8 3.0 0.3 5.0 9 6.3 1.0 2.3 27.4 58.3 G29 506 29 71.3 0.7 73.7 18.6 0.6 2.7 1.0 1.7 0.3 5.7 4 4.0 1.0 1.7 26.7 61.6 G30 820 13 67.7 3.0 70.7 47.0 0.6 3.3 1.3 1.8 0.7 1.3 8 7.0 1.0 0.0 30.0 61.1 G31 538 28 68.0 2.3 71.0 33.6 1.3 3.5 1.2 2.0 0.0 9.0 7 5.7 1.0 2.7 26.7 51.8 G32 431 40 63.7 2.3 66.0 33.0 0.9 3.2 1.5 2.0 0.0 7.3 6 2.3 1.0 1.0 24.7 46.7 G33 755 20 65.7 0.7 68.0 39.3 0.6 3.2 1.2 2.7 0.0 7.3 8 3.0 1.0 2.0 23.0 61.7 G34 398 41 65.0 1.3 65.7 38.0 0.9 3.7 1.5 2.5 0.0 7.0 8 7.3 1.2 1.7 30.7 53.3 G35 900 11 65.0 2.0 66.3 45.6 0.9 3.5 1.3 2.2 0.0 4.0 9 4.3 1.0 0.3 28.0 70.0 G36 789 17 65.3 1.0 67.3 36.6 0.6 2.7 1.5 2.3 0.0 5.3 6 3.7 1.0 2.0 27.7 62.0 G37 469 36 64.3 2.0 65.3 39.3 0.9 2.5 1.5 2.2 0.3 6.3 6 1.3 1.0 1.7 29.4 65.3 G38 505 30 69.0 1.3 71.0 26.6 0.7 2.3 1.2 2.0 0.0 6.0 5 0.3 1.0 0.0 31.7 65.9 G39 615 26 65.0 2.7 66.3 31.6 0.8 3.0 1.5 2.5 0.3 4.3 6 5.0 1.0 1.7 26.7 59.3 G40 469 35 66.7 1.3 67.7 34.3 0.9 2.3 1.3 1.8 0.3 8.7 6 2.7 1.0 0.7 30.7 51.9 G41 1000 5 64.0 0.0 65.3 48.6 0.9 3.5 1.3 1.7 0.0 3.3 9 2.7 1.0 1.3 24.0 68.3 G43 690 23 55.0 2.7 64.3 34.0 1.0 2.7 1.7 1.8 0.3 5.7 6 2.7 1.0 1.0 29.4 63.3 G44 762 19 65.9 1.6 57.7 47.3 0.8 3.0 1.7 2.5 0.0 1.3 9 2.7 1.0 1.3 27.0 68.9 G45 938 9 72.3 2.7 67.4 38.1 0.8 3.4 1.0 1.6 0.2 .2 69 4.8 1.0 1.4 28.4 61.4 MEAN 696 66.5 1.8 68.3 37.2 0.8 3.1 1.5 2.1 0.4 .67 6 4.1 1.0 1.5 28.7 61.1 P 0.73 0.0 0.1 0.0 0.3 0.3 0.0 0.1 0.3 0.3 0.3 0.0 0.5 0.2 0.3 0.8 LSD 695.8 3.1 1.9 3.7 19.6 0.4 0.9 27.5 0.6 1.6 .8 32 4.4 0.1 2.4 7.1 0.4 CV 61.8 2.9 63.8 3.3 32.3 31.3 18.7 0.7 27.5 2 46.726.4 65.7 6.0 102.0 15.3 31.3 Min 324.3 55.0 0.0 57.7 18.6 0.5 2.2 1.0 1.5 0.0 5.7 4 0.3 1.0 0.0 23.0 43.0 Max 1146 72.3 3.3 75.0 56.6 1.3 4.0 2.2 3.0 2.3 8.31 0 8.3 1.2 4.0 35.4 74.5 LSD = Least significant difference, CV = coeffitc oief nvariation, Min = minimum, Max = maximum, GYg =ra in yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi l interval (days), DS = days to silking (days), EeHa r= height (cm), EPP = ears per plant (#), LB =f lbelaight disease (1-5), GLS = gray leaf spot deis (e1a-s5), GT = grain texture (1-5), HC = husk cover, PH = plant height (cm), RL = roloodt ging (#), Rust = rust disease (1-5), SL = sotedmgi nl g (#), SWT = 100 seed weight (g), SH = shge pllienrcentage, # = numb er. 196 Appendix 10 Mean performance combined across tawros yaend across optimal and low pH environment0 f1o1r /212 and 2012/13 seasons Genotypes GY AD ASI DS PH EH EPP RL SL GLS LB MSV USRT GT SH% VIG RE HC SWT G1 2060.1 75.9 2.2 77.9 143.0 58.3 1.0 2.3 1.5 1.02 .3 1.1 1.6 2.0 70.3 2.7 0.9 1.2 27.2 G2 1368.4 73.4 1.5 75.2 136.5 57.6 0.9 3.0 1.2 1.02 .4 1.1 1.7 2.8 58.7 2.8 1.1 1.0 23.5 G3 2240.9 72.4 1.4 73.9 136.4 59.7 1.1 2.2 1.3 1.02 .0 1.1 1.4 2.2 77.3 2.4 0.9 1.4 25.9 G4 2159.8 73.6 2.4 75.9 136.8 57.8 1.0 2.0 0.8 1.02 .1 1.1 1.7 2.8 72.1 2.3 1.3 1.5 24.6 G5 2610.7 77.9 1.1 79.1 151.0 61.0 1.0 1.1 1.2 1.02 .0 0.9 1.6 2.3 69.1 1.9 0.9 1.5 26.9 G6 2222.2 73.2 1.3 74.6 155.8 66.7 1.1 2.5 1.0 1.02 .3 1.1 1.6 2.7 69.2 2.4 1.2 1.3 25.5 G7 1454.7 73.0 2.3 75.2 273.4 58.6 1.0 1.3 1.0 1.02 .1 1.1 1.5 2.5 71.6 2.3 1.4 1.1 29.2 G8 2739.4 74.7 1.8 76.5 152.8 66.0 1.1 1.5 1.6 1.02 .5 1.0 1.4 2.0 70.1 2.4 0.9 1.4 26.3 G9 2293.9 75.3 1.7 76.8 148.6 63.4 1.0 2.6 1.0 1.02 .1 1.0 1.7 2.5 70.3 2.2 0.9 0.9 26.3 G10 2529.0 77.0 2.1 79.1 147.9 64.8 1.6 1.8 1.0 1.01.9 1.1 1.7 3.0 73.5 2.1 1.2 0.8 26.6 G11 1628.9 76.3 1.4 77.6 133.0 52.6 1.0 2.7 1.2 1.02.0 1.1 1.7 3.2 74.3 2.8 0.3 1.2 22.3 G12 2121.4 73.9 2.5 76.2 145.7 59.6 1.1 3.0 1.2 1.02.2 1.1 1.8 2.0 71.2 2.6 1.1 1.1 22.5 G13 2511.7 74.6 2.5 77.0 142.1 62.1 1.1 1.7 1.0 1.02.1 1.2 1.5 3.3 74.8 2.6 1.5 1.4 25.8 G14 2108.0 74.2 1.9 76.1 159.2 60.0 1.0 2.0 1.3 1.02.1 1.1 1.4 2.5 68.3 2.1 0.7 1.3 27.5 G15 2244.0 71.5 2.8 74.2 138.2 56.1 1.0 1.4 1.0 1.02.4 1.0 1.8 2.5 71.2 2.6 1.2 1.8 24.3 G16 2131.7 69.6 2.2 71.7 136.8 58.3 1.0 1.7 1.2 1.02.0 1.1 1.7 3.3 71.1 2.2 1.6 2.6 27.3 G17 2158.6 73.8 2.5 76.3 143.6 61.7 1.1 1.8 0.9 1.02.1 1.0 1.5 2.3 73.8 2.6 0.9 1.6 24.1 G18 2600.8 73.9 1.1 75.1 148.4 64.2 1.1 2.0 1.2 1.02.2 1.0 1.6 2.3 73.3 2.2 0.7 1.4 26.5 G19 2223.4 74.3 1.2 75.5 142.0 60.9 1.0 1.7 1.2 1.02.3 1.1 1.6 1.8 75.0 2.6 0.9 2.1 25.3 G20 2848.8 75.6 1.9 77.4 146.5 65.9 1.0 1.7 0.8 1.01.9 1.1 1.6 2.7 72.9 2.0 0.9 1.8 28.1 G21 2350.0 74.1 1.8 75.9 150.8 64.6 1.1 1.7 1.3 1.02.3 1.0 1.7 2.8 74.4 2.3 0.9 1.4 26.2 G22 2389.2 71.7 2.8 74.4 146.1 61.3 1.1 2.1 1.3 1.02.2 1.1 1.6 2.3 72.4 2.4 1.1 1.4 25.4 G23 2028.7 70.5 2.4 72.8 131.5 52.3 1.0 3.2 1.6 1.02.3 1.1 1.8 2.5 70.1 2.8 1.6 1.1 23.7 G24 2662.0 69.9 2.3 72.2 146.0 60.7 1.1 1.3 1.2 1.02.5 0.9 1.6 2.0 74.6 2.4 0.7 1.5 25.7 G25 2473.3 72.5 0.8 73.4 146.1 63.1 1.1 1.7 1.1 1.02.1 1.1 1.7 2.3 73.9 2.3 1.2 1.4 26.2 G26 2615.8 75.8 2.5 78.1 148.1 62.9 1.0 1.0 1.1 1.02.1 1.1 1.5 1.7 69.6 2.3 0.9 1.3 26.2 G27 2934.4 76.7 1.9 78.6 158.2 70.4 1.2 1.8 1.0 1.02.2 1.1 1.5 2.3 74.1 1.9 0.9 1.3 26.4 G28 2698.0 72.7 2.1 74.8 149.0 65.7 1.1 2.3 1.4 1.02.1 1.1 1.7 2.3 72.7 2.1 0.7 0.9 25.1 197 Appendix 10 continued G29 2270.6 76.8 1.6 78.3 145.5 60.1 1.0 1.4 1.2 1.02.1 1.1 1.5 2.3 71.7 2.6 0.8 1.7 25.4 G30 2554.8 74.3 2.0 76.3 147.8 66.8 1.0 2.0 0.5 1.02.1 1.0 1.6 2.0 73.4 2.1 0.9 1.4 25.8 G31 2037.4 73.0 2.1 75.1 140.1 63.4 1.1 2.4 1.2 1.02.1 1.1 1.7 2.3 68.6 2.5 1.1 1.3 23.0 G32 2271.4 71.7 2.1 73.7 145.8 62.5 1.1 2.0 1.3 1.02.2 1.0 1.8 2.8 71.4 2.3 0.9 1.4 24.5 G33 2424.7 71.9 2.1 74.1 146.9 63.6 1.1 2.0 1.3 1.02.0 1.1 1.6 1.5 72.0 2.4 1.0 1.1 23.3 G34 2177.4 70.3 1.8 72.0 144.2 61.0 1.2 3.6 0.9 1.02.4 1.1 1.7 2.0 74.2 2.7 0.8 1.2 22.9 G35 2303.4 74.3 2.1 76.3 146.1 61.1 1.0 2.6 1.1 1.02.1 0.9 1.6 2.5 75.1 2.5 0.9 2.3 27.2 G36 2509.3 72.1 2.3 74.4 147.7 63.7 1.0 1.7 1.2 1.02.1 1.1 1.9 1.5 73.0 2.2 1.2 1.3 25.6 G37 1450.0 71.5 1.8 73.3 134.9 51.8 1.1 1.5 1.0 1.01.9 1.1 1.6 2.2 70.1 2.8 0.7 0.8 24.9 G38 1000.1 76.1 1.4 77.3 142.3 59.5 0.9 1.0 0.9 1.02.1 0.9 1.4 1.8 66.0 2.2 1.4 0.9 27.0 G39 2518.1 71.4 1.8 73.3 137.9 55.3 1.1 2.1 1.2 1.02.1 1.1 1.7 3.0 74.3 2.4 0.9 1.5 26.5 G40 1573.2 76.7 1.7 78.3 140.0 56.7 1.0 1.1 0.5 1.01.9 1.1 1.4 3.0 69.6 2.5 0.7 1.3 26.9 G41 2153.2 67.6 2.0 69.7 132.4 51.7 1.0 2.0 1.1 0.72.0 1.0 1.8 2.3 76.1 2.7 0.8 1.2 25.0 G42 2109.8 73.8 2.0 75.8 138.9 57.6 1.0 1.3 1.2 0.02.1 1.0 1.6 2.3 71.1 2.4 1.4 0.8 25.2 G43 2632.7 70.4 3.1 73.4 146.3 60.1 1.0 1.7 0.9 1.02.0 1.0 1.8 1.7 69.5 2.5 0.9 1.4 26.5 G44 2617.6 72.3 2.2 74.6 146.2 61.5 1.0 1.9 1.0 1.02.1 1.0 1.6 2.8 70.1 2.6 0.7 1.3 26.7 G45 2764.9 75.7 1.9 77.7 148.6 64.8 1.0 1.6 0.9 1.01.9 1.1 1.6 2.7 71.5 2.1 0.8 1.0 28.2 Mean 2261.7 73.5 2.0 70.5 147.2 60.8 1.1 1.9 1.1 .7 71 2.1 1.0 1.6 2.4 71.7 2.4 1.0 1.3 25.7 Min 1000.1 67.6 0.8 131.5 51.7 0.9 1.0 0.5 0.0 1.90 .9 1.4 1.5 58.7 1.9 0.3 0.8 22.3 22.3 Max 2934.4 77.9 3.1 273.4 70.4 1.6 3.6 1.6 1.0 2.51 .2 1.9 3.3 77.3 2.8 1.6 2.6 29.2 29.2 Min = minimum, Max = maximum, GY = grain yield (khga- 1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (da)y, sP)H, = plant height (cm), EH = ear height (cm), EPP = ears per plan),t R(#L = root lodging (#), SL = stem lodging (#),G =L gSray leaf spot disease (1-5), LB = leaf bligishet adse (1-5), MSV = maize streak virus disease (1-5), Rust = rust disea5s)e, (G1T- = grain texture (1-5), SH = shelling percaegnet, VIG = vigour, RE = rotten ears (#), HC = hucsokv er, SWT = 100 seed weight (g), # = number.. 198 Appendix 11 Genotypes used in genotype x envirot ninmternactions and stability analysis Genotype code Pedigree G1 DT-WSTR SYNTHETIC-B G2 EVD-W 99 STR QPM CO-B G3 IAR-FLINT-Q-B G4 IWD C3 SYN F2-B G5 MULTICOB EARLY DT -B G6 SYN DTE STY-W-B G7 OBA SUPER1(9021-18(IITA))-B G8 LPHpop 9 G9 POP66 SR/DMR -LSRY/DMR-LSRY G10 POP66 SR/TZUTSR-WSGY/T G11 LPHpop 7 G12 TZE-YDT STR C4-B G13 VPO0721 G14 TZE E -WPOP X LD(SET2)-B G15 VPO5173 G16 VPO717 G17 VPO5187 G18 VPO52 G19 VPO630 G20 LPHpop 21 G21 VPO741 G22 VPO739 G23 LPHpop 18 G24 LPHpop 16 G25 VPO743 G26 VPO744 G27 LOW N POOL C3-B G28 LPHpop 10 G29 VPO76 G30 LPHpop 3 G31 LPHpop 20 G32 LPHpop 2 G33 VPO710 G34 VPO712 G35 LPHpop 8 G36 OBATANPA/TZLCOMP4C3F2/TZLCOMP4C3F2- B G37 VPO738 G38 LPHpop 13 G39 LPHpop 15 G40 VPO86 G41 VPO96 G42 VPO97 G43 ZM309 Check 1 G44 ZM523 Check 2 G45 ZM721 Check 3 199 Appendix 12 Estimated specific combining abilfietyc tesf of 12 inbred lines for grain yield and agmroicn otraits across low pH and optimal environments Cross Pedigree GY AD ASI DS EH EPP LB GLS GT MSV PH RL Rust SWT SH% SL VIG S12 CZL999601/CML481 -0.646 1.74 -1.71 3.57 2.14 -5.37 -0.40 -0.48 0.303. 07 -5.04 0.47 0.22 -0.25 -0.76 -0.72 0.00 S13 CZL999601/CML359 -0.148 -0.18 -1.98 1.38 3.51 3.08 0.11 0.86 -0.205. 11- 5.06 -0.80 0.44 1.76 -4.32 -0.68 -0.30 S14 CZL999601/CML144 0.943 2.14 -0.62 2.54 2.66 -4.24 -0.50 -0.04 0.400 .01- 0.81 -0.20 0.03 3.53 -5.75 1.20 -0.30 S15 CZL999601/CML161 -0.13 -0.57 -1.20 1.13 -0.74 2.61 -0.20 -0.07 - 0.1-60.11 -3.89 -0.10 0.09 0.05 -6.84 -0.28 -0.10 S16 CZL999601/CML172 -0.449 -1.87 1.26 -1.86 0.78 2.84 -0.10 0.18 0.100 .05- 3.17 -0.80 -0.18 -2.09 41*** 0.24 0.10 S17 CZL999601/CML448 -0.097 -2.54 -0.45 -1.17 4.05 -4.36 0.82 0.16 0.203. 02 6.48 -0.50 0.07 -0.54 -3.28 0.16 0.99 S18 CZL999601/CML312 0.401 -1.61 0.99 -2.47 -6.67 -1.62 -0.10 0.06 - 0.209.03 -1.25 0.87 -0.46 -0.23 2.73 0.48 0.05 S19 CZL999601/CZL130-23 0.212 0.61 1.15 -0.28 -0.63 0.04 -0.20 -0.10 -0.3-05. 02 -0.94 0.89 -0.06 -0.52 -1.11 0.41 -0.70 S110 CZL999601/CML288 -0.361 0.37 0.95 -2.40 -0.92 2.66 0.69 0.15 0.37 .01- 0 -3.07 0.36 0.01 -2.47 -9.89 -0.22 0.47 S111 CZL999601/CML202 0.668 -0.16 -0.97 0.66 -4.62 1.74 -0.10 -0.61 - 0.009.11 -5.65 0.03 0.14 2.48 -1.08 -0.18 0.30 S23 CML481/CML359 -0.387 0.32 0.10 -0.07 -0.71 2.29 0.65 0.21 0.23 .07- 0 0.22 -1.30 0.17 -3.67 2.34 2.38 0.60 S24 CML481/CML144 -0.316 0.65 -1.17 1.90 0.21 -4.37 0.88 -0.35 0.040 .03- 0.86 1.72 0.21 -1.48 10.49 0.15 0.18 S25 CML481/CML161 0.487 -2.44 -0.75 -1.37 -2.68 4.74 0.06 -0.11 - 0.009.04 1.21 1.71 -0.07 -1.61 -3.99 0.11 0.30 S26 CML481/CML172 -0.219 0.18 1.79 -0.59 -2.86 4.14 -0.10 -0.13 - 0.0-03.07 -1.19 -1.20 -0.12 -3.49 -6.57 -0.59 0.85 S27 CML481/CML448 0.403 2.14 -0.46 1.28 -6.09 -4.51 -0.50 0.06 0.38. 00 0 -16.8 2.82 0.13 1.37 7.76 -0.34 -0.60 S28 CML481/CML312 -0.079 2.44 1.16 1.53 -1.31 -1.56 0.24 0.24 -0.04.1 2 0 -3.39 -1.50 -0.01 -0.15 -7.11 0.54 -0.10 S29 CML481/CZL130-23 0.595 -0.41 2.57 -2.02 3.59 3.75 -0.30 0.08 -0.84.0 2 0 11.30 -0.30 -0.17 3.69 -2.41 -0.43 -0.60 S210 CML481/CML288 0.703 -2.18 -1.97 -1.07 4.27 0.47 -0.60 0.16 -0.1-0. 14 7.43 -1.60 -0.28 3.19 1.55 -0.61 -0.50 S211 CML481/CML202 -0.3 -0.84 -1.89 1.58 0.80 3.20 0.02 0.24 0.32 0.0-40.09 -0.60 0.15 -0.93 6.21 -0.23 -0.20 S34 CML359/CML144 -0.593 1.96 -2.34 5.21 -2.52 -3.54 -0.20 -0.40 0.0-00.15 -1.07 -0.60 -0.07 -0.79 6.07 -0.59 0.62 S35 CML359/CML161 0.189 0.43 0.03 0.36 6.07 -2.01 -0.30 0.13 -0.19. 25- 0 5.18 -0.70 -0.01 5.57 -2.19 0.03 -0.20 S36 CML359/CML172 0.404 0.82 0.85 -0.85 -3.31 -1.38 -0.50 0.04 -0.4-02. 20 -5.69 -0.20 -0.11 2.14 -7.08 -0.22 -0.50 S37 CML359/CML448 0.236 2.42 0.32 1.48 -2.98 3.30 0.38 -0.26 0.06 0.04 -7.54 1.31 -0.09 -2.64 7.71 -0.19 0.10 200 Appendix 12 continue d S38 CML359/CML312 0.012 -1.50 0.20 -2.05 6.58 -0.35 0.44 0.30 -0.18 -0.06 6.24 -1.40 0.05 -1.27 -1.26 -0.42 -0.10 S39 CML359/CZL130-23 -0.032 0.78 -1.53 0.37 3.20 1.97 -0.20 -0.46 -0.24 0.01 3.66 -0.40 -0.22 -2.15 2.12 -0.50 0.04 S310 CML359/CML288 0.144 -6.71 3.90 -5.26 -6.93 1.77 -0.30 -0.41 0.91 0.92 0.69 0.19 -0.02 -2.59 -4.89 0.27 -0.10 S311 CML359/CML202 -0.249 -0.47 -0.10 -0.77 -0.86 -3.04 0.05 0.19 0.20 -0.08 -2.13 0.03 0.04 -0.62 -0.16 -0.19 -0.40 S45 CML144/CML161 -0.234 -1.63 -2.90 0.34 4.24 -1.81 0.09 0.51 0.08 0.07 -2.76 -0.80 0.08 -0.77 -0.35 0.02 0.15 S46 CML144/CML172 -0.218 -1.94 -0.42 -1.42 -5.95 3.23 -0.20 -0.47 0.26 -0.10 -2.22 -0.20 -0.30 1.19 -5.11 -0.01 0.28 S47 CML144/CML448 -0.156 -2.53 -2.04 -0.36 -5.21 3.55 0.56 0.01 0.03 0.14 -7.19 -1.10 0.11 -1.29 6.61 -0.20 0.26 S48 CML144/CML312 0.431 -1.71 1.01 -2.99 2.33 -0.65 -0.30 -0.37 -0.44 -0.07 0.66 2.19 0.14 3.51 -0.84 -0.10 -0.30 S49 CML144/ZL130-23 -0.475 3.80 2.79 0.85 -6.46 1.10 -0.10 -0.03 -0.67 0.00 -13.0 1.09 -0.13 -1.89 -8.34 -0.18 -0.10 S410 CML144/CML288 0.208 -1.12 -0.31 -0.88 -0.11 -3.61 -0.10 0.55 0.25 0.17 -6.27 2.35 0.22 -1.86 -3.25 -0.14 -0.30 S411 CML144/CML202 0.731 -1.30 2.60 -4.03 7.70 7.40 -0.20 0.62 -0.16 0.01 18.64 -0.30 -0.44 3.00 0.15 -0.26 -0.30 S56 CML161/CML172 0.671 0.43 0.29 -0.63 0.46 -0.44 -0.30 0.28 0.12 0.47 -4.05 1.41 -0.19 0.26 2.91 0.06 -0.50 S57 CML161/CML448 -0.359 -0.93 0.00 -0.15 -6.13 -1.99 0.29 -0.52 0.03 -0.13 2.43 -1.70 -0.39 1.24 -2.43 -0.24 0.64 S58 CML161/CML312 -0.12 0.40 1.43 -0.99 -8.88 0.63 0.18 -0.01 -0.17 0.22 -11.0 0.28 -0.14 0.30 2.04 -0.36 0.19 S59 CML161/CZL130-23 0.094 -2.56 -1.92 -1.41 2.94 -2.72 0.31 0.17 -0.78 -0.16 4.70 -0.60 -0.02 -3.97 4.02 0.67 -0.20 S510 CML161/CML288 -0.557 4.03 4.14 0.90 -6.64 -2.65 -0.40 -0.14 -0.52 -0.27 -2.01 -1.20 0.06 2.53 8.84 0.38 0.11 S511 CML161/CML202 -0.558 1.41 0.86 0.20 4.13 3.99 0.00 -0.12 2.5*** -0.09 -0.89 0.11 0.35 -4.81 -8.44 -0.13 -0.10 S67 CML172/CML448 -0.185 -1.34 -0.68 -1.45 -0.41 -2.13 0.16 0.01 0.50 -0.02 6.91 3.29 0.51 -0.25 -15.15 0.40 -0.10 S68 CML172/CML312 -0.301 0.57 -0.26 0.30 1.31 0.76 0.77 0.02 0.17 -0.12 -0.73 0.32 0.15 -0.41 2.33 -0.51 0.24 S69 CML172/CZL130-23 0.521 -0.89 -1.03 0.46 2.25 -2.58 -0.20 0.64 -0.10 0.23 5.94 -0.80 0.33 2.40 1.80 0.08 0.13 S610 CML172/CML288 -0.107 1.54 -0.04 2.03 7.96 -2.56 -0.30 -0.22 -0.29 -0.05 8.02 -0.40 -0.05 1.04 -0.67 0.46 -0.50 S611 CML172/CML202 -0.013 1.59 -0.10 1.27 -4.40 -1.29 0.83 -0.43 -0.08 -0.03 -7.51 0.31 -0.08 -2.44 -9.69 0.06 0.58 S78 CML448/CML312 0.225 -1.04 -1.79 0.78 -0.07 -0.95 -0.50 0.16 0.16 -0.04 0.09 -0.80 0.12 1.21 -0.48 -0.03 -0.40 S79 CML448/CZL130-23 0.13 0.25 0.60 -0.14 -4.27 -4.33 0.00 -0.11 -0.49 0.08 -1.35 0.12 -0.49 5.04 7.08 0.12 -0.40 S710 CML448/CML288 -0.06 3.53 1.27 1.06 12.49 16.6*** -0.40 -0.09 0.13 -0.14 13.18 -1.00 0.03 -0.50 -12.54 0.27 -0.20 S711 CML448/CML202 0.039 1.99 3.10 -1.62 8.12 -2.87 -0.50 0.38 -0.75 0.04 12.83 -0.50 0.16 1.22 0.83 -0.13 0.00 S89 CML312/CZL130-23 -0.161 -1.09 -2.88 1.94 3.81 5.02 -0.20 0.34 2.2*** 0.04 4.26 -0.80 0.32 -0.81 -4.94 -0.01 -0.30 S810 CML312/CML288 -0.233 3.09 2.21 0.73 -2.63 -1.53 0.41 -0.41 0.06 -0.13 -6.77 -0.60 -0.05 -4.23 0.89 -0.19 -0.30 201 Appendix 12 continue d S811 CML312/CML202 0.145 2.06 0.62 2.00 1.79 -0.30 -0.60 -0.44 -0.76 0.00 3.61 -1.20 0.02 5.8** 2.21 0.52 -0.10 S910 CZL130-23/CML288 -0.254 -0.79 -3.50 2.16 5.32 -4.91 0.59 -0.34 -0.53 -0.17 -0.65 0.90 0.35 1.55 7.76 0.07 1.08 S911 CZL130-23/CML202 -0.65 1.10 1.94 0.24 -3.31 -3.71 0.34 -0.10 -0.95 0.01 -5.54 1.15 -0.30 2.71 3.13 -0.11 0.24 S112 CZL999601/CML539 -0.393 2.07 2.58 -1.09 0.43 2.62 -0.10 -0.11 -0.29 0.07 4.32 -0.30 -0.29 -1.72 -10.72 -0.41 -0.40 S212 CML481/CML539 -0.241 -1.60 2.33 -4.75 2.65 -2.79 0.03 0.08 -0.18 0.00 5.53 -0.50 -0.23 3.32 -7.50 -0.25 0.07 S312 CML359/CML539 0.422 2.12 0.54 0.20 -2.05 -2.09 -0.20 -0.19 -0.12 -0.07 -4.61 1.75 -0.17 4.25 1.66 0.12 0.18 S412 CML144/CML539 -0.322 1.66 3.39 -1.16 3.10 2.94 0.20 -0.04 0.21 -0.03 11.57 -0.90 0.14 -3.13 0.34 0.10 -0.20 S512 CML161/CML539 0.518 1.41 0.03 1.61 7.23 -0.36 0.30 -0.12 -0.83 0.21 11.10 1.11 0.26 1.20 6.42 -0.26 -0.30 S612 CML172/CML539 -0.105 0.92 -1.66 2.73 4.17 -0.59 -0.10 0.08 -0.23 -0.07 -2.64 -2.60 0.04 1.65 -3.79 0.04 -0.60 S712 CML448/CML539 -0.176 -1.94 0.14 0.29 0.50 -2.32 -0.30 0.22 -0.29 0.00 -9.00 -0.20 -0.16 -4.86 3.89 0.19 -0.30 S812 CML312/CML539 -0.32 -1.62 -2.68 1.21 3.74 0.54 -0.20 0.12 -0.70 0.01 8.29 2.32 -0.13 -3.76 4.45 0.06 1.04 S912 ZL130-23/CML539 0.019 -0.79 1.82 -2.17 -6.44 6.37 0.06 -0.09 2.8*** -0.03 -8.34 -1.00 0.38 -6.05 -9.11 -0.13 0.76 S1011 CML288/CML202 0.053 -2.46 -3.12 0.05 -4.41 -3.54 0.19 0.48 -0.09 -0.05 -3.79 0.17 -0.23 -6.1** 2.35 -0.07 0.36 S1012 CML288/CML539 0.465 0.69 -3.55 2.68 -8.39 -2.73 0.24 0.27 -0.17 -0.14 -6.75 0.17 -0.05 9. *** 9.86 -0.20 0.00 S1112 CML202/CML539 0.165 -3.70 -1.41 0.06 -8.15 -3.56 0.20 0.25 0.08 0.03 -13.1 1.50 0.42 1.81 2.36 1.24 -0.30 GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daEyHs) =, ear height (cm), EPP = ears per plant (#), =L Ble af blight disease (1-5), GLS = gray leaf spot dis(e1a-5s)e, GT = grain texture (1-5), MSV = maize stre vaikrus disease (1-5), PH = plant height (cm), RrLo o=t lodging (#), Rust = rust disease (1-5), SWT = 100 seed weight (g), SH =li nshge plercentage, SL = stem lodging (#), VIG = vri,g #o u= numbe.r 202 Appendix 13 Mean performance of diallel cross pnryo gaecross optimal and low pH environments in 220 11/1 Pedigree GY AD ASI DS EH LB EPP GLS GT MSV PH RL Rust SWT SH% SL VIG 1 CZL999601/CML481 3809 69.0 0.0 69.0 91.1 1.5 1.0 1.2 1.8 1.2 163.5 6.0 1.2 33.2 80.0 0.7 2.0 2 CZL999601/CML359 5210 73.0 0.3 73.3 82.7 2.0 1.0 2.7 2.3 1.0 178.1 5.7 1.8 38.0 82.7 0.3 2.0 3 CZL999601/CML144 6955 74.0 0.7 74.7 89.4 1.0 1.2 1.8 2.8 1.0 173.3 3.0 1.2 42.7 81.9 4.7 1.0 4 CZL999601/CML161 3740 74.0 2.0 76.0 102.8 2.3 1.0 1.0 2.0 1.0 195.9 4.3 1.2 38.4 79.1 0.0 2.2 5 CZL999601/CML172 3624 69.0 1.0 70.0 94.3 2.3 1.0 2.3 2.2 1.0 189.5 6.7 1.0 31.2 87.7 2.3 2.2 6 CZL999601/CML448 2663 70.0 2.0 72.0 89.3 4.0 1.0 1.0 2.2 1.0 174.0 3.0 1.0 33.1 80.5 2.0 3.8 7 CZL999601/CML312 4024 69.3 0.7 70.0 92.7 1.5 1.0 1.2 2.0 1.0 195.2 5.0 1.2 35.5 86.5 2.7 2.7 8 CZL999601/ZL130-2 3 5099 72.0 0.0 72.0 95.6 1.5 1.2 1.2 1.8 1.0 191.7 8.3 1.0 36.3 83.9 2.3 1.0 9 CZL999601/CML288 3779 74.0 0.0 74.0 90.6 3.7 1.1 1.0 2.3 1.0 169.4 10.3 1.2 36.8 83.4 0.7 2.8 10 CZL999601/CML20 2 5386 74.0 1.0 75.0 71.3 2.0 1.0 1.0 2.3 1.2 163.8 3.7 1.0 38.6 85.3 1.0 2.3 11 CZL999601/CML53 9 3939 72.0 0.7 72.7 92.3 1.7 1.0 1.0 2.0 1.2 180.3 4.0 1.0 33.3 77.3 0.0 1.0 12 CML481/CML359 3128 71.7 0.7 72.3 99.2 3.2 1.0 1.3 2.3 1.2 188.0 4.7 1.5 31.7 85.0 9.3 2.8 13 CML481/CML144 3628 74.0 1.3 75.3 96.4 3.7 1.1 1.0 1.8 1.0 179.9 1.3 1.0 28.8 79.1 3.3 2.3 14 CML481/CML161 3566 70.0 2.0 72.0 89.2 2.3 1.2 1.0 2.2 1.5 177.8 7.3 1.0 28.0 84.0 2.7 2.7 15 CML481/CML172 2349 67.3 1.0 68.3 100.6 3.3 1.0 1.2 2.0 1.0 184.8 12.0 1.0 25.2 81.9 0.3 3.0 16 CML481/CML448 4976 76.0 0.3 76.3 70.8 1.3 1.0 1.0 2.3 1.0 124.2 1.3 1.2 38.2 79.9 1.0 1.3 17 CML481/CML312 2509 75.3 0.7 76.0 82.2 3.8 1.0 1.0 1.8 1.5 180.0 16.0 1.2 28.2 81.8 3.7 2.3 18 CML481/ZL130-23 3612 73.3 2.3 75.7 90.8 1.7 1.1 1.2 2.2 1.0 188.0 2.0 1.0 45.2 83.6 0.3 1.0 19 CML481/CML288 4676 72.0 0.3 72.3 92.0 1.5 1.0 1.3 2.2 1.0 181.8 5.7 1.0 43.8 86.8 0.7 1.2 20 CML481/CML202 2702 72.0 1.7 73.7 90.5 2.2 1.2 1.8 2.7 1.0 171.5 1.7 1.2 33.8 87.4 0.3 1.7 21 CML481/CML539 2509 70.0 0.7 70.7 79.3 2.0 1.0 1.0 2.3 1.0 175.6 5.7 1.0 34.9 82.3 0.7 1.7 22 CML359/CML144 2484 72.0 2.0 74.0 90.9 1.5 1.0 1.0 2.0 1.0 190.7 2.7 1.0 34.8 80.9 0.3 2.8 23 CML359/CML161 2535 73.0 2.3 75.3 96.5 1.7 1.0 1.2 2.5 1.0 188.7 2.0 1.2 42.2 70.8 0.3 2.0 24 CML359/CML172 4640 73.3 2.7 76.0 81.1 1.3 1.3 1.3 1.8 1.0 168.9 5.0 1.0 37.5 78.3 1.7 1.0 25 CML359/CML448 3073 76.0 0.0 76.0 93.4 3.0 1.0 1.0 2.2 1.5 185.9 2.7 1.0 39.2 82.7 1.3 2.3 26 CML359/CML312 2344 69.0 2.7 71.7 84.3 3.2 1.2 5 1. 2.3 1.0 180.0 10.7 1.0 25.3 82.5 1.0 2.3 27 CML359/ZL130-23 2722 74.0 0.0 74.0 109.7 1.3 1.0 1.0 2.5 1.2 197.3 1.7 1.0 32.9 74.7 0.3 2.8 28 CML359/CML288 3621 70.0 2.0 72.0 81.8 1.5 1.0 2 1. 1.8 1.2 173.2 3.7 1.0 38.0 81.4 1.7 2.0 203 Appendix 13 continue d 29 CML359/CML202 3170 70.3 1.7 72.0 108.6 2.0 1.0 2.2 2.0 1.0 194.3 3.7 1.0 39.7 81.7 0.3 1.2 30 CML359/CML539 3353 76.0 2.0 78.0 97.4 1.5 1.0 1.2 2.2 1.2 188.6 10.0 1.0 34.0 77.6 1.3 1.8 31 CML144/CML161 2142 71.0 0.0 71.0 101.0 2.0 1.0 1.7 2.7 1.0 187.7 4.3 1.0 32.5 83.1 0.7 2.3 32 CML144/CML172 2379 70.0 1.0 71.0 94.3 1.7 1.1 1.0 2.2 1.0 184.5 3.0 1.2 37.4 90.1 1.3 2.8 33 CML144/CML448 2513 71.3 0.3 71.7 74.1 2.3 1.0 1.7 2.0 1.5 151.3 2.0 1.0 31.8 86.3 0.0 2.3 34 CML144/CML312 3280 74.0 0.3 74.3 86.8 1.7 1.0 1.0 2.2 1.0 172.0 2.7 1.3 46.4 83.1 0.7 2.3 35 CML144/ZL130-23 1835 76.0 0.7 76.7 95.8 1.5 1.2 1.0 1.7 1.2 174.3 10.0 1.2 33.7 74.5 0.0 2.2 36 CML144/CML288 3210 70.0 2.0 72.0 88.5 1.7 1.1 2.3 2.5 1.0 178.6 7.3 1.8 31.5 87.2 1.0 1.5 37 CML144/CML202 4172 70.0 1.0 71.0 91.0 1.5 1.1 2.3 2.3 1.0 184.3 5.7 1.0 38.0 83.4 0.3 1.2 38 CML144/CML539 2513 70.0 1.7 71.7 94.7 1.7 1.1 1.0 2.0 1.0 183.4 2.0 1.0 31.9 83.8 1.3 1.2 39 CML161/CML172 5026 75.0 1.3 76.3 98.1 1.2 1.0 1.3 2.7 1.0 194.8 4.0 1.0 40.7 83.8 0.3 1.0 40 CML161/CML448 757 72.0 3.7 75.7 88.1 4.3 0.9 1.0 2.2 1.0 177.9 8.0 1.0 27.9 70.0 0.0 4.5 41 CML161/CML312 1805 72.0 0.3 72.3 85.5 1.3 1.0 1.0 2.5 1.2 170.7 2.3 1.2 29.8 75.1 0.3 3.0 42 CML161/ZL130-23 2652 75.3 0.7 76.0 85.7 2.7 1.0 1.5 1.8 1.0 167.3 3.7 1.0 28.6 80.7 0.7 1.8 43 CML161/CML288 1381 74.0 0.7 74.7 84.3 1.3 1.0 1.0 2.2 1.0 180.2 3.0 1.0 29.5 79.4 0.0 3.5 44 CML161/CML202 1556 76.0 0.7 76.7 122.2 2.7 1.1 1.0 2.0 1.0 216.9 1.0 1.0 31.5 72.5 0.3 2.8 45 CML161/CML539 4713 74.0 2.0 76.0 88.2 1.5 1.1 1.0 1.8 1.0 174.1 9.7 1.2 34.0 82.3 0.0 2.3 46 CML172/CML448 2514 69.3 0.7 70.0 85.9 2.8 1.1 1.0 1.8 1.2 179.9 12.3 1.2 29.7 74.3 0.3 1.5 47 CML172/CML312 2365 69.7 1.3 71.0 65.2 4.0 1.0 2.0 2.7 1.0 162.9 9.0 1.0 28.9 82.7 0.3 3.7 48 CML172/ZL130-23 5218 74.0 2.0 76.0 85.9 2.0 1.1 1.3 2.3 1.0 180.4 5.0 1.5 37.0 80.7 1.3 2.0 49 CML172/CML288 3251 74.0 0.7 74.7 81.8 1.5 1.1 1.2 1.8 1.5 169.4 6.3 1.2 37.9 79.5 0.7 1.8 50 CML172/CML202 3649 76.0 1.3 77.3 104.2 3.2 1.0 1.5 2.2 1.0 194.3 5.3 1.2 29.0 81.4 1.3 2.8 51 CML172/CML539 2775 78.0 2.0 80.0 95.0 1.3 1.0 1.0 2.0 1.0 173.3 0.3 1.0 45.7 77.6 0.3 1.3 52 CML448/CML312 2448 75.7 1.7 77.3 95.0 1.5 0.9 1.0 2.5 1.0 192.3 4.3 1.0 39.7 68.2 1.3 2.7 53 CML448/ZL130-23 2882 74.0 2.0 76.0 109.2 2.0 1.0 1.2 1.7 1.3 199.4 2.7 1.2 42.6 80.2 0.0 1.2 54 CML448/CML288 2732 78.0 0.7 78.7 104.0 1.5 1.1 1.7 2.7 1.0 191.2 3.3 1.0 39.0 84.9 0.7 1.7 55 CML448/CML202 2239 77.0 0.0 77.0 92.9 1.5 1.1 1.0 2.2 1.0 186.3 2.7 1.0 31.7 75.6 0.3 2.8 56 CML448/CML539 3243 69.0 3.7 72.7 93.6 2.0 1.1 1.0 1.7 1.0 182.0 5.3 1.0 44.8 84.9 0.3 2.2 57 CML312/ZL130-23 2448 76.3 1.0 77.3 86.9 1.7 1.1 1.0 3.0 1.3 178.4 5.3 1.0 32.0 79.4 0.3 2.8 204 Appendix 13 58 CML312/CML288 3737 69.0 3.0 72.0 111.3 2.3 1.0 1.2 2.2 1.2 187.1 6.0 1.2 33.7 87.6 0.0 1.8 59 CML312/CML202 3820 72.0 0.7 72.7 97.6 1.5 1.0 1.0 2.5 1.0 182.5 1.0 1.0 42.2 86.4 1.0 2.0 60 CML312/CML539 1622 69.0 1.3 70.3 84.0 2.5 1.0 1.2 2.2 1.2 172.7 10.3 1.0 27.7 75.2 0.7 4.2 61 ZL130-23/CML288 1456 74.0 0.0 74.0 104.8 3.8 1.0 1.3 1.7 1.0 190.3 8.0 1.3 24.5 77.0 0.0 4.2 62 ZL130-23/CML202 2323 74.3 1.7 76.0 106.2 1.3 1.1 1.0 1.5 1.0 194.7 4.7 1.0 33.8 80.6 0.3 2.3 63 ZL130-23/CML539 2792 72.0 0.7 72.7 89.0 2.3 1.0 1.0 2.5 1.0 170.9 2.0 1.0 28.3 78.7 0.3 3.2 64 CML288/CML202 2653 69.7 0.7 70.3 90.7 2.7 1.3 2.7 2.2 1.2 186.3 2.7 1.0 25.4 80.9 0.7 2.7 65 CML288/CML539 3677 75.7 0.3 76.0 83.4 1.7 1.1 1.5 2.3 1.0 181.0 2.3 1.0 48.9 81.4 0.3 1.7 66 CML202/CML539 3942 71.0 0.7 71.7 102.9 1.3 1.3 1.0 1.8 1.0 193.2 3.0 1.3 34.7 82.1 3.0 1.0 67 MH27 4744 74.7 1.3 76.0 86.9 1.0 1.1 1.0 2.0 1.0 192.3 5.7 1.0 34.2 77 0.7 1.3 68 MH26 6018 74.3 1.7 76.0 111.3 1.0 1.0 1.2 2.0 1.0 199.4 7.0 1.0 41.2 83 0.0 1.2 Mean 3205.2 72.6 1.2 73.8 92.0 2.1 1.1 1.3 2.2 1.11 80.9 5.0 1.1 34.9 81.0 1.0 2.2 LSD 1650 0.8 1.4 1.3 328.2 0.8 0.2 0.5 0.8 0.4 38.76.7 1.1 10.2 10.5 3.8 1.1 MSE 4E+06 0.3 0.8 0.6 29.3 0.3 0.0 0.1 0.3 0.1 .15 74 17.2 0.5 39.5 42.3 5.6 0.5 P 0.001 0.001 0.00 0 0.63 0.0 0.17 0 0.4 0.2 0.750 .00 0.08 0.001 0.05 0.35 0 CV (%) 31.9 0.7 74.7 0.8 19.7 24.3 12.4 22.8 23.7 21.3 13.2 82.9 0.2 18.0 8.0 230.1 30.8 LSD = Least significant differen cMeS, E = Mean square error, CV =coefficient of variationG,Y = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi l interval (days), DS = days to silking (days), EeHa r= h eight (cm), LB = leaf blight disease (1-5)P, E=P ears per plant (#), GLS = gray leaf spot dies e(1a-s5), GT = grain texture (1-5), MSV = maize streak virus disease (1-5), PH = plhaenitg ht (cm), RL = root lodging (#), Rust = ruset adsise (1-5), SWT = 100 seed weight (g), SH = shge plleinrcentage, SL = stem lodging (#), VIG = vigour, # = numb. e r 205 Appendix 14 Mean performances of diallel crosse pnrioegs across three low pH environments in 2012 Pedigree GY AD ASI DS EH EPP LB GLS GT MSV PH RL URST SWT SH% SL VIG 1 CZL999601/CML481 894 83.4 2.9 86.3 46.8 0.5 2.3 1.3 2.2 1.0 110 0.5 2.4 22.7 62.0 2.2 4.3 2 CZL999601/CML359 678 79.3 3.1 82.4 47.3 0.5 2.3 2.5 1.5 1.0 117 0.7 2.5 27.0 66.4 1.8 4.0 3 CZL999601/CML144 853 82.7 2.6 85.2 46.1 0.4 1.3 1.7 2.2 1.0 111 0.7 2.2 18.4 68.2 1.3 5.0 4 CZL999601/CML161 657 79.2 2.4 81.7 38.3 0.5 2.0 2.0 2.1 1.0 97 0.5 .3 2 23.6 78.5 1.8 4.7 5 CZL999601/CML172 787 78.8 3.8 82.6 42.5 0.7 2.3 1.9 2.2 1.0 108 0.7 2.2 24.6 74.0 1.3 4.7 6 CZL999601/CML448 953 77.4 3.0 80.4 43.8 0.5 2.7 2.1 2.6 1.0 113 0.5 2.3 18.9 73.6 1.0 5.0 7 CZL999601/CML312 867 81.2 3.0 84.2 30.2 0.5 2.7 2.1 1.9 1.1 97 0.8 .7 1 20.9 65.8 1.8 4.7 8 CZL999601/ZL130-2 3 945 82.9 3.1 86.0 41.5 0.5 2.3 1.8 2.1 1.0 104 0.5 2.3 24.0 70.2 1.5 4.3 9 CZL999601/CML288 821 79.6 2.2 81.8 37.5 0.5 2.7 2.5 2.3 1.2 102 0.5 2.0 23.5 65.4 1.0 5.0 10 CZL999601/CML20 2 971 79.6 2.6 82.1 40.4 0.5 2.3 1.3 1.9 1.0 104 0.5 2.6 20.4 69.5 2.3 5.0 11 CZL999601/CML53 9 664 83.9 2.4 86.3 35.6 0.6 2.3 1.6 1.8 1.0 105 0.7 1.9 22.7 61.7 2.0 4.7 12 CML481/CML359 631 80.4 3.7 84.1 37.1 0.5 3.0 2.1 2.3 1.0 106 0.8 2.1 16.3 59.0 2.2 4.7 13 CML481/CML144 670 80.8 2.7 83.4 40.7 0.5 3.0 1.5 1.6 1.0 107 0.5 2.3 22.8 73.0 1.3 4.3 14 CML481/CML161 835 78.7 2.4 81.1 39.5 0.5 3.0 1.8 1.7 1.0 105 0.5 1.9 20.5 73.4 3.0 4.7 15 CML481/CML172 695 82.4 3.9 86.3 34.9 0.4 1.3 1.9 2.1 1.0 98 0.5 .1 2 20.9 58.2 1.8 5.0 16 CML481/CML448 752 81.8 2.1 83.9 38.6 0.5 1.7 1.8 2.2 1.0 99 0.5 .2 2 23.4 66.2 1.5 4.0 17 CML481/CML312 651 84.1 3.0 87.1 39.9 0.4 1.7 2.3 2.6 1.0 110 0.5 2.2 19.7 59.5 1.8 4.3 18 CML481/ZL130-23 958 83.0 3.2 86.2 44.4 0.5 2.0 2.0 1.9 1.1 110 0.5 2.0 17.2 65.0 1.5 4.3 19 CML481/CML288 999 77.8 3.0 80.8 47.8 0.5 1.3 2.3 1.7 1.0 113 0.5 1.6 21.6 66.8 1.5 4.3 20 CML481/CML202 872 79.3 2.9 82.2 43.6 0.6 2.7 2.1 2.1 1.0 111 0.5 2.3 23.0 73.4 1.2 4.3 21 CML481/CML539 721 80.6 2.8 83.3 41.6 0.5 2.7 1.8 2.3 1.0 105 0.7 1.8 21.0 56.3 1.3 4.7 22 CML359/CML144 643 82.1 3.8 85.9 41.5 0.5 1.0 1.4 1.7 1.0 110 0.5 2.0 21.3 61.6 1.0 5.0 23 CML359/CML161 686 80.3 3.0 83.3 50.7 0.4 1.7 2.1 2.1 1.0 115 0.5 2.0 23.9 62.4 2.2 4.7 24 CML359/CML172 628 81.0 2.7 83.7 41.5 0.7 1.3 2.1 1.9 1.0 104 0.5 2.2 18.8 62.2 0.5 4.7 25 CML359/CML448 1008 81.9 3.4 85.3 32.6 0.8 1.7 1.3 2.2 1.0 93 0.5 2.0 21.8 70.3 1.3 4.7 26 CML359/CML312 685 81.3 3.3 84.7 47.8 0.8 2.0 2.2 2.2 1.2 113 0.7 2.4 19.9 65.1 0.8 4.7 27 CML359/ZL130-23 815 81.9 2.7 84.6 43.1 0.6 2.0 1.3 2.2 1.2 112 0.5 2.0 20.6 66.5 0.5 4.0 28 CML359/CML288 638 81.6 2.3 84.5 35.0 0.7 1.0 1.6 1.8 3.5 99 0.8 .3 2 20.8 58.4 1.6 4.7 206 Appendix 14 continued 29 CML359/CML202 569 79.8 3.2 83.0 35.4 0.5 2.0 1.8 2.6 1.0 105 0.72 .3 16.9 53.7 1.7 4.7 30 CML359/CML539 804 82.3 2.4 84.8 30.8 0.5 1.7 1.3 2.1 1.0 93 0.7 .0 2 22.1 72.3 1.7 5.0 31 CML144/CML161 704 78.7 2.3 81.0 47.6 0.5 2.3 2.5 2.0 1.3 102 0.5 2.2 26.7 64.8 1.2 5.0 32 CML144/CML172 821 78.7 3.2 81.9 34.6 0.6 1.7 1.6 2.0 1.0 100 0.5 1.8 26.8 64.4 1.3 4.3 33 CML144/CML448 825 77.6 3.2 80.8 38.2 0.5 2.7 1.5 1.8 1.0 103 0.5 2.3 27.5 72.2 1.7 5.0 34 CML144/CML312 725 80.6 3.0 83.6 41.5 0.6 1.0 1.5 1.8 1.0 102 0.5 2.3 21.5 74.7 1.5 4.3 35 CML144/ZL130-23 827 85.7 2.8 88.4 38.0 0.5 1.7 2.0 1.7 1.0 102 0.5 2.0 21.0 83.2 1.8 4.3 36 CML144/CML288 712 79.7 2.8 82.4 39.4 0.5 1.3 2.2 1.7 1.0 102 0.5 2.1 20.0 63.6 1.7 4.7 37 CML144/CML202 736 81.6 3.1 84.7 44.7 0.4 1.3 2.3 1.7 1.0 114 0.5 1.9 26.2 72.3 3.2 5.0 38 CML144/CML539 821 82.8 2.8 85.6 43.0 0.6 2.3 1.6 2.4 1.0 121 0.8 2.3 22.3 70.9 1.3 4.3 39 CML161/CML172 773 80.3 2.3 82.7 44.3 0.5 3.0 2.5 2.3 2.0 95 0.5 .0 2 22.5 71.2 2.0 5.0 40 CML161/CML448 608 79.3 2.3 81.7 34.3 0.7 1.0 1.0 2.4 1.0 107 0.5 1.5 27.2 57.9 1.0 4.0 41 CML161/CML312 435 83.0 3.0 86.0 33.3 0.4 4.0 2.0 2.3 1.5 94 0.5 .0 2 22.2 65.2 0.5 5.0 42 CML161/ZL130-23 581 78.0 2.3 80.3 51.0 0.5 3.0 2.0 1.9 1.0 116 1.0 2.3 18.0 80.5 2.5 5.0 43 CML161/CML288 487 84.7 3.7 88.3 36.1 0.6 2.0 2.0 1.6 1.0 105 0.5 2.0 27.6 61.0 2.5 4.0 44 CML161/CML202 458 80.7 2.3 83.0 46.4 0.6 2.0 2.0 2.3 1.0 106 0.5 2.8 20.5 57.5 0.5 4.0 45 CML161/CML539 594 82.7 2.7 85.3 50.1 0.8 4.0 1.5 1.6 1.5 117 0.5 2.5 22.2 68.9 0.5 4.0 46 CML172/CML448 464 79.3 2.7 82.0 44.1 0.4 2.0 2.0 2.6 1.0 114 1.0 3.0 25.6 35.9 3.0 5.0 47 CML172/CML312 468 83.7 2.7 86.3 55.0 0.6 3.0 1.8 2.2 1.0 112 0.5 2.8 23.8 72.8 1.0 4.0 48 CML172/ZL130-23 811 80.3 3.0 83.3 51.4 0.7 2.0 3.0 2.4 1.5 115 0.5 2.8 25.1 70.1 0.5 5.0 49 CML172/CML288 636 81.0 3.7 84.7 57.7 0.6 2.0 2.0 1.6 1.0 123 0.5 2.0 22.8 64.3 2.0 4.0 50 CML172/CML202 478 81.0 2.7 83.7 35.6 0.5 4.0 1.5 2.0 1.0 95 0.5 .3 2 20.0 50.0 1.5 5.0 51 CML172/CML539 896 80.3 3.0 83.3 45.0 0.6 3.0 2.0 2.1 1.0 100 0.5 2.5 20.3 65.4 1.0 4.0 52 CML448/CML312 509 79.3 2.7 82.0 38.7 0.4 1.0 2.0 2.4 1.0 101 0.5 2.5 21.0 62.5 0.5 4.0 53 CML448/ZL130-23 554 81.7 2.7 84.3 30.3 0.5 2.0 1.5 2.0 1.0 97 1.0 .5 1 25.5 64.9 2.0 5.0 54 CML448/CML288 373 83.0 2.7 85.7 47.3 0.4 1.0 1.5 1.9 1.0 112 0.5 2.0 32.4 69.8 1.0 5.0 55 CML448/CML202 508 81.3 3.3 84.7 44.8 0.5 1.0 2.5 1.0 1.0 110 0.5 2.5 23.8 67.3 0.5 4.0 56 CML448/CML539 426 79.3 3.7 83.0 36.0 0.5 1.0 1.8 1.9 1.0 87 0.5 .0 2 12.7 62.5 1.0 4.0 57 CML312/ZL130-23 472 81.0 3.0 84.0 50.3 0.6 2.0 2.5 1.6 1.0 114 1.0 3.0 25.0 63.0 0.5 4.0 207 Appendix 14 continue d 58 CML312/CML288 360 85.0 2.3 87.3 38.3 0.7 3.0 1.5 2.2 1.0 103 1.0 2.0 17.1 61.4 1.0 5.0 59 CML312/CML202 589 84.7 3.3 88.0 43.4 0.5 1.0 1.5 1.4 1.0 115 0.5 2.5 26.2 61.1 2.5 5.0 60 CML312/CML539 583 81.7 3.0 84.7 45.3 0.6 1.0 1.8 1.6 1.0 116 0.5 2.3 17.6 70.6 3.5 5.0 61 ZL130-23/CML288 824 79.7 3.0 82.7 45.7 0.4 2.0 1.5 1.9 1.0 105 1.0 2.5 28.7 67.8 1.5 5.0 62 ZL130-23/CML202 266 82.7 3.3 86.0 35.7 0.6 4.0 2.0 1.6 1.0 98 0.5 .0 2 25.0 61.6 2.5 5.0 63 ZL130-23/CML539 799 81.0 3.3 84.3 39.0 0.5 2.0 1.5 2.9 1.0 104 0.5 3.0 20.2 58.9 1.5 5.0 64 CML288/CML202 1041 77.7 2.7 80.3 34.1 0.4 2.0 2.3 1.5 1.0 101 0.51.8 16.5 65.6 2.5 5.0 65 CML288/CML539 829 80.0 2.3 82.3 28.7 0.4 3.0 2.0 1.9 1.0 95 0.5 .0 2 27.4 70.7 3.5 5.0 66 CML202/CML539 847 77.3 3.3 80.7 28.7 0.4 3.0 1.5 2.2 1.0 88 0.5 .5 2 17.9 64.6 2.0 5.0 67 MH27 639.5 77.7 1.3 79.0 57.0 1.0 1.7 2.0 2.0 1.0 166.30 .0 1.0 33.6 59.9 2.2 5.0 68 MH26 879.3 78.7 1.0 79.7 65.0 1.0 1.3 2.0 2.0 1.3 171.0 .0 1.7 29.9 63.2 3.5 4.5 Mean 701.6 80.9 2.9 83.8 41.7 0.5 2.1 1.9 2.0 1.11 07 0.6 2.2 22.5 65.6 1.6 4.6 LSD 293.3 3.1 0.7 0.4 12.6 0.3 0.5 0.8 0.8 0.7 3 19. 0.3 0.7 6.9 57.8 1.6 0.6 MSE 1001 3.7 0.5 4.0 61.4 0.0 0.2 0.2 0.3 0.2 145 0.0 0.2 18.3 1295. 1.0 0.1 P 0.001 0.00 0.001 0.001 0.001 0.001 0.001 0.00.10 010 0.001 0.001 0.001 0.001 0.001 0.005 0.001 10 .00 CV (%) 45.2 2.4 24.6 2.4 19.0 39.3 26.9 0.5 26.0 .54 2 11.4 36.4 19.4 19.0 52.1 63.8 7.7 LSD = Least significant differen cMeS, E = Mean square error, CV =coefficient of variatio nG,Y = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi l interval (days), DS = days to silking (days), EeHa r= h eight (cm), EPP = ears per plant (#), LB =f lbelaight disease (1-5), GLS = gray leaf spot deis (e1a-s5), GT = grain texture (1-5), MSV = maize streak virus disease (1-5), PH = plhaenitg ht (cm), RL = root lodging (#), Rust = ruset adsise (1-5), SWT = 100 seed weight (g), SH = shge plleinrcentage, SL = stem lodging (#), VIG = vigour, # = numb. e r 208 Appendix 15 Mean performance of the diallel crorosgs epnies under optimal conditions in 2011/12 Pedigree GY AD ASI DS EH LB EPP GLS GT MSV PH RL RUST SWT SH% SL VIG 1 CZL999601/CML481 3809 69.0 0.0 69.0 91.1 1.5 1.0 1.2 1.8 1.2 163.5 6.0 1.2 33.2 80.0 0.7 2.0 2 CZL999601/CML359 5210 73.0 0.3 73.3 82.7 2.0 1.0 2.7 2.3 1.0 178.1 5.7 1.8 38.0 82.7 0.3 2.0 3 CZL999601/CML144 6955 74.0 0.7 74.7 89.4 1.0 1.2 1.8 2.8 1.0 173.3 3.0 1.2 42.7 81.9 4.7 1.0 4 CZL999601/CML161 3740 74.0 2.0 76.0 102.8 2.3 1.0 1.0 2.0 1.0 195.9 4.3 1.2 38.4 79.1 0.0 2.2 5 CZL999601/CML172 3624 69.0 1.0 70.0 94.3 2.3 1.0 2.3 2.2 1.0 189.5 6.7 1.0 31.2 87.7 2.3 2.2 6 CZL999601/CML448 2663 70.0 2.0 72.0 89.3 4.0 1.0 1.0 2.2 1.0 174.0 3.0 1.0 33.1 80.5 2.0 3.8 7 CZL999601/CML312 4024 69.3 0.7 70.0 92.7 1.5 1.0 1.2 2.0 1.0 195.2 5.0 1.2 35.5 86.5 2.7 2.7 8 CZL999601/ZL130-2 3 5099 72.0 0.0 72.0 95.6 1.5 1.2 1.2 1.8 1.0 191.7 8.3 1.0 36.3 83.9 2.3 1.0 9 CZL999601/CML288 3779 74.0 0.0 74.0 90.6 3.7 1.1 1.0 2.3 1.0 169.4 10.3 1.2 36.8 83.4 0.7 2.8 10 CZL999601/CML202 5386 74.0 1.0 75.0 71.3 2.0 1.0 1.0 2.3 1.2 163.8 3.7 1.0 38.6 85.3 1.0 2.3 11 CZL999601/CML539 3939 72.0 0.7 72.7 92.3 1.7 1.0 1.0 2.0 1.2 180.3 4.0 1.0 33.3 77.3 0.0 1.0 12 CML481/CML359 3128 71.7 0.7 72.3 99.2 3.2 1.0 1.3 2.3 1.2 188.0 4.7 1.5 31.7 85.0 9.3 2.8 13 CML481/CML144 3628 74.0 1.3 75.3 96.4 3.7 1.1 1.0 1.8 1.0 179.9 1.3 1.0 28.8 79.1 3.3 2.3 14 CML481/CML161 3566 70.0 2.0 72.0 89.2 2.3 1.2 1.0 2.2 1.5 177.8 7.3 1.0 28.0 84.0 2.7 2.7 15 CML481/CML172 2349 67.3 1.0 68.3 100.6 3.3 1.0 1.2 2.0 1.0 184.8 12.0 1.0 25.2 81.9 0.3 3.0 16 CML481/CML448 4976 76.0 0.3 76.3 70.8 1.3 1.0 1.0 2.3 1.0 124.2 1.3 1.2 38.2 79.9 1.0 1.3 17 CML481/CML312 2509 75.3 0.7 76.0 82.2 3.8 1.0 1.0 1.8 1.5 180.0 16.0 1.2 28.2 81.8 3.7 2.3 18 CML481/ZL130-23 3612 73.3 2.3 75.7 90.8 1.7 1.1 1.2 2.2 1.0 188.0 2.0 1.0 45.2 83.6 0.3 1.0 19 CML481/CML288 4676 72.0 0.3 72.3 92.0 1.5 1.0 1.3 2.2 1.0 181.8 5.7 1.0 43.8 86.8 0.7 1.2 20 CML481/CML202 2702 72.0 1.7 73.7 90.5 2.2 1.2 1.8 2.7 1.0 171.5 1.7 1.2 33.8 87.4 0.3 1.7 21 CML481/CML539 2509 70.0 0.7 70.7 79.3 2.0 1.0 1.0 2.3 1.0 175.6 5.7 1.0 34.9 82.3 0.7 1.7 22 CML359/CML144 2484 72.0 2.0 74.0 90.9 1.5 1.0 1.0 2.0 1.0 190.7 2.7 1.0 34.8 80.9 0.3 2.8 23 CML359/CML161 2535 73.0 2.3 75.3 96.5 1.7 1.0 1.2 2.5 1.0 188.7 2.0 1.2 42.2 70.8 0.3 2.0 24 CML359/CML172 4640 73.3 2.7 76.0 81.1 1.3 1.3 1.3 1.8 1.0 168.9 5.0 1.0 37.5 78.3 1.7 1.0 25 CML359/CML448 3073 76.0 0.0 76.0 93.4 3.0 1.0 1.0 2.2 1.5 185.9 2.7 1.0 39.2 82.7 1.3 2.3 26 CML359/CML312 2344 69.0 2.7 71.7 84.3 3.2 1.2 1.5 2.3 1.0 180.0 10.7 1.0 25.3 82.5 1.0 2.3 27 CML359/ZL130-23 2722 74.0 0.0 74.0 109.7 1.3 1.0 1.0 2.5 1.2 197.3 1.7 1.0 32.9 74.7 0.3 2.8 28 CML359/CML288 3621 70.0 2.0 72.0 81.8 1.5 1.0 1.2 1.8 1.2 173.2 3.7 1.0 38.0 81.4 1.7 2.0 209 29 CML359/CML202 3170 70.3 1.7 72.0 108.6 2.0 1.0 2.2 2.0 1.0 194.3 3.7 1.0 39.7 81.7 0.3 1.2 30 CML359/CML539 3353 76.0 2.0 78.0 97.4 1.5 1.0 1.2 2.2 1.2 188.6 10.0 1.0 34.0 77.6 1.3 1.8 31 CML144/CML161 2142 71.0 0.0 71.0 101.0 2.0 1.0 1.7 2.7 1.0 187.7 4.3 1.0 32.5 83.1 0.7 2.3 32 CML144/CML172 2379 70.0 1.0 71.0 94.3 1.7 1.1 1.0 2.2 1.0 184.5 3.0 1.2 37.4 90.1 1.3 2.8 33 CML144/CML448 2513 71.3 0.3 71.7 74.1 2.3 1.0 1.7 2.0 1.5 151.3 2.0 1.0 31.8 86.3 0.0 2.3 34 CML144/CML312 3280 74.0 0.3 74.3 86.8 1.7 1.0 1.0 2.2 1.0 172.0 2.7 1.3 46.4 83.1 0.7 2.3 35 CML144/ZL130-23 1835 76.0 0.7 76.7 95.8 1.5 1.2 1.0 1.7 1.2 174.3 10.0 1.2 33.7 74.5 0.0 2.2 36 CML144/CML288 3210 70.0 2.0 72.0 88.5 1.7 1.1 2.3 2.5 1.0 178.6 7.3 1.8 31.5 87.2 1.0 1.5 37 CML144/CML202 4172 70.0 1.0 71.0 91.0 1.5 1.1 2.3 2.3 1.0 184.3 5.7 1.0 38.0 83.4 0.3 1.2 38 CML144/CML539 2513 70.0 1.7 71.7 94.7 1.7 1.1 1.0 2.0 1.0 183.4 2.0 1.0 31.9 83.8 1.3 1.2 39 CML161/CML172 5026 75.0 1.3 76.3 98.1 1.2 1.0 1.3 2.7 1.0 194.8 4.0 1.0 40.7 83.8 0.3 1.0 40 CML161/CML448 757 72.0 3.7 75.7 88.1 4.3 0.9 1.0 2.2 1.0 177.9 8.0 1.0 27.9 70.0 0.0 4.5 41 CML161/CML312 1805 72.0 0.3 72.3 85.5 1.3 1.0 1.0 2.5 1.2 170.7 2.3 1.2 29.8 75.1 0.3 3.0 42 CML161/ZL130-23 2652 75.3 0.7 76.0 85.7 2.7 1.0 1.5 1.8 1.0 167.3 3.7 1.0 28.6 80.7 0.7 1.8 43 CML161/CML288 1381 74.0 0.7 74.7 84.3 1.3 1.0 1.0 2.2 1.0 180.2 3.0 1.0 29.5 79.4 0.0 3.5 44 CML161/CML202 1556 76.0 0.7 76.7 122.2 2.7 1.1 1.0 2.0 1.0 216.9 1.0 1.0 31.5 72.5 0.3 2.8 45 CML161/CML539 4713 74.0 2.0 76.0 88.2 1.5 1.1 1.0 1.8 1.0 174.1 9.7 1.2 34.0 82.3 0.0 2.3 46 CML172/CML448 2514 69.3 0.7 70.0 85.9 2.8 1.1 1.0 1.8 1.2 179.9 12.3 1.2 29.7 74.3 0.3 1.5 47 CML172/CML312 2365 69.7 1.3 71.0 65.2 4.0 1.0 2.0 2.7 1.0 162.9 9.0 1.0 28.9 82.7 0.3 3.7 48 CML172/ZL130-23 5218 74.0 2.0 76.0 85.9 2.0 1.1 1.3 2.3 1.0 180.4 5.0 1.5 37.0 80.7 1.3 2.0 49 CML172/CML288 3251 74.0 0.7 74.7 81.8 1.5 1.1 1.2 1.8 1.5 169.4 6.3 1.2 37.9 79.5 0.7 1.8 50 CML172/CML202 3649 76.0 1.3 77.3 104.2 3.2 1.0 1.5 2.2 1.0 194.3 5.3 1.2 29.0 81.4 1.3 2.8 51 CML172/CML539 2775 78.0 2.0 80.0 95.0 1.3 1.0 1.0 2.0 1.0 173.3 0.3 1.0 45.7 77.6 0.3 1.3 52 CML448/CML312 2448 75.7 1.7 77.3 95.0 1.5 0.9 1.0 2.5 1.0 192.3 4.3 1.0 39.7 68.2 1.3 2.7 53 CML448/ZL130-23 2882 74.0 2.0 76.0 109.2 2.0 1.0 1.2 1.7 1.3 199.4 2.7 1.2 42.6 80.2 0.0 1.2 54 CML448/CML288 2732 78.0 0.7 78.7 104.0 1.5 1.1 1.7 2.7 1.0 191.2 3.3 1.0 39.0 84.9 0.7 1.7 55 CML448/CML202 2239 77.0 0.0 77.0 92.9 1.5 1.1 1.0 2.2 1.0 186.3 2.7 1.0 31.7 75.6 0.3 2.8 56 CML448/CML539 3243 69.0 3.7 72.7 93.6 2.0 1.1 1.0 1.7 1.0 182.0 5.3 1.0 44.8 84.9 0.3 2.2 57 CML312/ZL130-23 2448 76.3 1.0 77.3 86.9 1.7 1.1 1.0 3.0 1.3 178.4 5.3 1.0 32.0 79.4 0.3 2.8 58 CML312/CML288 3737 69.0 3.0 72.0 111.3 2.3 1.0 1.2 2.2 1.2 187.1 6.0 1.2 33.7 87.6 0.0 1.8 210 59 CML312/CML202 3820 72.0 0.7 72.7 97.6 1.5 1.0 1.0 2.5 1.0 182.5 1.0 1.0 42.2 86.4 1.0 2.0 60 CML312/CML539 1622 69.0 1.3 70.3 84.0 2.5 1.0 1.2 2.2 1.2 172.7 10.3 1.0 27.7 75.2 0.7 4.2 61 ZL130-23/CML288 1456 74.0 0.0 74.0 104.8 3.8 1.0 1.3 1.7 1.0 190.3 8.0 1.3 24.5 77.0 0.0 4.2 62 ZL130-23/CML202 2323 74.3 1.7 76.0 106.2 1.3 1.1 1.0 1.5 1.0 194.7 4.7 1.0 33.8 80.6 0.3 2.3 63 ZL130-23/CML539 2792 72.0 0.7 72.7 89.0 2.3 1.0 1.0 2.5 1.0 170.9 2.0 1.0 28.3 78.7 0.3 3.2 64 CML288/CML202 2653 69.7 0.7 70.3 90.7 2.7 1.3 2.7 2.2 1.2 186.3 2.7 1.0 25.4 80.9 0.7 2.7 65 CML288/CML539 3677 75.7 0.3 76.0 83.4 1.7 1.1 1.5 2.3 1.0 181.0 2.3 1.0 48.9 81.4 0.3 1.7 66 CML202/CML539 3942 71.0 0.7 71.7 102.9 1.3 1.3 1.0 1.8 1.0 193.2 3.0 1.3 34.7 82.1 3.0 1.0 67 MH27 4744 74.7 1.3 76.0 86.9 1.0 1.1 1.0 2.0 1.0 192.3 5.7 1.0 34.2 77 0.7 1.3 68 MH26 6018 74.3 1.7 76.0 111.3 1.0 1.0 1.2 2.0 1.0 199.4 7.0 1.0 41.2 83 0.0 1.2 Mean 3205.2 72.6 1.2 73.8 92.0 2.1 1.1 1.3 2.2 1.118 0.9 5.0 1.1 34.9 81.0 1.0 2.2 LSD 1650 0.8 1.4 1.3 328.2 0.8 0.2 0.5 0.8 0.4 38.76.7 1.1 10.2 10.5 3.8 1.1 MSE 4E+06 0.3 0.8 0.6 29.3 0.3 0.0 0.1 0.3 0.1 15 74. 17.2 0.5 39.5 42.3 5.6 0.5 P 0.001 0.001 0.001 0 0.63 0.0 0.17 0 0.36 0.16 5 0.7 0 0.08 0.001 0.05 0.35 0 CV (%) 31.9 0.7 74.7 0.8 19.7 24.3 12.4 22.8 23.7 1.32 13.2 82.9 0.2 18.0 8.0 230.1 30.8 SE 1021 0.5 0.9 1.1 18.1 0.5 0.1 0.3 0.5 0.2 24.0 .2 4 25.5 6.3 6.5 2.4 0.7 MIN 757.0 67.3 0.0 68.3 65.2 1.0 0.9 1.0 1.5 1.0 4.122 0.3 1.0 24.5 68.2 0.0 1.0 MAX 6955.0 78.0 3.7 80.0 122.2 4.3 1.3 2.7 3.0 1.251 6.9 16.0 1.8 48.9 90.1 9.3 4.5 LSD = Least significant differen cMeS, E = Mean square error, CV =coefficient of variation, SE = error, MIN = minimu, MAX = maximum,G Y = grain yield (kg h-a1), AD = days to anthesis (days), ASI = anthesis-sgi liknitnerval (days), DS = days to silking (days),= E eHa r height (cm), LB = leaf blight disease (1E-P5)P, = ears per plant (#), GLS = gray leaf spot disease (1-5), GT = grain textu-r5e) ,( 1MSV = maize streak virus disease (1-5), PHl a=n pt height (cm), RL = root lodging (#), Rust =t rduisease (1-5), SWT = 100 seed weight (g), SH = shelling percentage, SL = stemgi nlogd (#), VIG = vigour, # = numb.e r 211 Appendix 16 Estimated general combining abiliteyc tesf ffor 12 inbred lines for grain yield and agmroinc otraits at low pH environments in 2011/12 Name GY AD ASI DS EH EPP GLS GT LB MSV PH RL RUST WS T SH% SL VIG 1 G1 CZL999601 0.12 -0.07 -0.37 0.13 -0.15 0.30 1 0.00.01 0.07 -0.06 1.06 0.02 -0.05 0.95 7.60* 0.050 .51- 2 G2 CML481 0.03 0.28 0.11 0.24 0.22 0.44 0.02 1- 0.10.03 -0.02 0.93 0.16 -0.20 -1.03 -3.22 -0.01 9- 0.8 3 G3 CML359 0.03 -0.61 0.16 -0.33 -0.85 1.62* -0 .080.09 -0.04 0.06 0.25 -0.18 0.37 -1.81* -5.18 0.030. 16* 4 G4 CML144 0.17 0.05 0.16 -0.16 0.00 -0.14 -0.040 .17- -0.21 -0.02 0.71 -0.01 -0.22 1.36 1.16 -0.050 .65- 5 G5 CML161 -0.19 -0.31 -0.23 -0.69 2.00 -0.75 0 .080.05 0.20 0.19** 0.23 -0.08 -0.15 0.92 0.41 0.050 .64- 6 G6 CML172 -0.04 -0.18 -0.38 0.52 3.51* -0.06 0 .160.10 0.08 0.07 0.79 -0.20 0.06 0.43 4.23 0.02 7- 0.5 7 G7 CML448 -0.09 -0.67 -0.52 -0.18 -2.28 -0.71 16-0 . 0.07 -0.32* -0.10* -2.10 -0.25 -0.11 1.51 -0.20-0 .05 -0.67 8 G8 CML312 -0.20 1.69** 0.67 1.10 0.98 -0.70 0.050 .04 -0.08 -0.02 1.13 -0.21 0.07 -0.97 -0.59 0.070 .81- 9 G9 ZL130-23 0.03 0.82 0.28 0.60 1.69 0.12 0.05 02 0. 0.05 0.02 1.20 0.01 0.05 0.42 -0.10 -0.01 -0.65 10 G10 CML288 0.11 -0.90 0.75 -1.86 -0.43 -0.11 1 0.1-0.09 0.05 0.03 -0.83 0.21 0.17 1.39 0.89 0.06 24 3. 11 G11 CML202 0.00 -0.42 -0.18 -0.12 -1.91 -0.75 040 . -0.13 0.07 -0.10* -0.62 0.34* 0.03 -0.56 -2.980 .1-2 -0.44 12 G12 CML539 0.03 0.32 -0.44 0.73 -2.78 0.74 - 0.204.14 0.10 -0.03 -2.74 0.19 -0.01 -2.62** -2.01 0-10 . -0.57 **P ≤0.01; *P≤0.05; GY = grain yield (kg h-1a), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daEyHs) =, ear height (cm), EPP = ears per plant (#), GLS = gray leaf spot disease (1-5), G gTr a=in texture (1-5), LB = leaf blight disease )(,1 M-5SV = maize streak virus disease (1-5), PH =n pt lhaeight (cm), RL = root lodging (#), Rust = rust disease (1-5), SWT = 1e0e0d sweight (g), SH = shelling percentage, SL = lsotdegming (#), VIG = vigour, # = numb. er 212 Appendix 17 Estimated specific combining abilfietyc tesf of 12 inbred lines for grain yield and agmroicn otraits across low pH environments 2011/12 Cross Pedigree GY AD ASI DS EH EPP LB GLS GT MSV HP RL SWT SH SL VIG Rust 1 S12 CZL999601/CML481 -0.09 2.45 0.77 1.73 5.68 .53- 1 -0.02 -0.66* 0.35 0.00 2.70 0.31 0.56 -6.96 04-0 . 0.60 0.41 2 S13 CZL999601/CML359 -0.31 -0.78 -2.28 1.64 7.20 7.76** 0.05 0.69* -0.53 -0.09 10.37 0.48 2.01 -182 .5 -0.05 -3.78 -0.07 3 S14 CZL999601/CML144 -0.03 1.89 -0.39 1.80 5.23 0.88- -0.28 -0.18 0.04 -0.01 4.25 -0.52 -1.24 -4.42 0.33 1.03 0.18 4 S15 CZL999601/CML161 -0.02 -1.19 -1.33 0.00 -4 .54 -0.22 -0.36 0.03 0.09 -0.22* -9.72 0.05 -0.66 1-7 .0 0.06 0.69 0.19 5 S16 CZL999601/CML172 -0.13 -1.76 0.48 -0.65 -1 .94 -0.73 -0.07 -0.14 -0.08 -0.09 0.94 0.00 1.61 5*6*. 17-0.07 0.62 -0.10 6 S17 CZL999601/CML448 0.36 -2.61 -1.04 -1.18 5.18 -0.16 0.49 0.36 0.09 0.07 9.28 -0.29 -3.26 -4.19 .17- 0 1.05 0.24 7 S18 CZL999601/CML312 0.37 -1.19 0.88 -1.79 -141**. 7 -0.22 0.26 0.15 -0.15 0.08 -10.29 0.35 -0.84 .34-5 0.21 0.85 -0.61 8 S19 CZL999601/CZL130-23 0.08 1.35 2.27 -0.40 2- 1.1 -1.16 -0.04 -0.10 0.12 -0.04 -3.80 -0.20 0.32 89- 1. -0.04 0.36 0.08 9 S110 CZL999601/CML288 -0.24 -0.27 -0.87 -0.39 89-2 . -0.94 0.13 0.50 0.35 0.09 -3.55 -0.90* -0.49 47-6 . -0.11 -2.86 -0.38 10 S111 CZL999601/CML202 0.30 -0.75 -1.38 -0.02 8 1.4 -0.22 -0.06 -0.59 0.00 0.07 -1.42 0.46 -0.02 0.35 -0.10 0.82 0.35 11 S23 CML481/CML359 -0.26 -0.01 1.24 -0.25 -3.29 .393 0.41 0.27 0.04 0.01 -0.72 0.51 -0.61 -2.03 0.3-52.73 -0.34 12 S24 CML481/CML144 -0.36 -0.34 -0.64 -0.20 -0.47 -1.09 0.59 -0.35 -0.07 -0.05 0.38 -0.16 -0.43 6.73 -0.27 0.74 0.49 13 S25 CML481/CML161 0.27 -2.09 -2.25 -0.22 -3.80 0.42- 0.18 -0.14 -0.42 0.00 -1.26 1.58** -0.71 7.38 -0.37 1.07 0.01 14 S26 CML481/CML172 0.07 1.56 1.89 1.23 -9.87* 12-1 . -0.54 -0.14 -0.25 0.01 -8.70 0.20 -0.49 -5.47 00 0. 1.34 -0.03 15 S27 CML481/CML448 -0.02 1.38 -0.07 -0.18 -0.30 0.46- 0.03 0.10 0.19 0.03 -4.59 0.08 0.50 4.29 - 0.100.44 0.22 16 S28 CML481/CML312 0.07 1.36 1.51 -0.12 -4.78 50-0 . -0.21 0.39 0.38 -0.04 -6.05 0.05 2.31 -1.81 0.11-0.76 0.04 17 S29 CML481/CZL130-23 0.34 0.24 2.24 -1.94 2.57 1.22- -0.17 0.06 -0.46 0.13 4.26 -0.34 -1.01 6.31 03 0. 0.91 -0.11 18 S210 CML481/CML288 0.15 -2.39 -3.45 1.39 6.96 155**. -0.50 0.24 -0.28 -0.09 7.80 -0.20 -2.97 -1.89 -0.38 -3.15 -0.73 19 S211 CML481/CML202 0.01 -1.32 -2.08 0.54 4.22 .38-0 0.14 0.15 0.38 0.03 5.04 -1.34** 1.90 10.37 70 .4 0.54 0.25 20 S34 CML359/CML144 -0.39 1.88 -0.36 3.60 1.37 22-2 . -0.34 -0.33 -0.13 -0.14 3.50 -0.32 -0.86 1.17 .12- 0 -2.64 -0.41 21 S35 CML359/CML161 0.21 0.47 0.70 0.13 8.49 - 1.61-0.42 0.21 -0.12 -0.06 9.20 0.25 3.40 2.42 0.45 .98- 2 -0.47 22 S36 CML359/CML172 -0.12 1.00 -0.38 0.14 -2.25 254* . -0.47 0.13 -0.46 -0.22* -2.58 -0.47 -4.15 -8 .57 -0.35 -3.05 -0.51 23 S37 CML359/CML448 0.23 2.38 0.76 2.29 -5.35 2-1 .3 0.10 -0.30 -0.02 -0.05 -10.58 0.41 -2.01 11.11 .29- 0 -2.95 -0.51 24 S38 CML359/CML312 0.11 -0.53 -2.42 2.23 6.62 22-1 . 0.03 0.33 0.17 0.04 6.63 -0.12 -1.16 -4.33 - 0.2-42.81 -0.28 25 S39 CML359/CZL130-23 0.15 0.89 -0.81 0.73 1.14 2.31- -0.10 -0.59 0.03 0.26 4.89 -0.67 0.09 7.17 32- 0. -3.64 -0.67 26 S310 CML359/CML288 0.12 -7.16** 3.22 -10.78 -74 .3 -2.10 1.20** -0.10 0.35 0.42** -11.04 0.13 -0.77- 3.43 0.11 30.47** 4.20** 213 Appendix 17 continue d 27 S311 CML359/CML202 -0.29 0.02 -0.02 0.11 -2.931 .60- -0.12 0.00 0.72* -0.05 0.27 -0.17 -2.69 -2.68 0.29 -3.18 -0.32 28 S45 CML144/CML161 -0.11 -1.86 -2.30 -0.04 4.52 .180 0.09 0.59 0.14 0.07 -4.48 -0.42 0.91 -1.06 0.00 1.16 0.28 29 S46 CML144/CML172 0.44 -1.99 -1.04 -0.59 -9.88*- 0.42 -0.13 -0.41 0.09 -0.14 -6.93 0.03 2.98 -5.36-0 .14 0.43 -0.35 30 S47 CML144/CML448 -0.09 -2.62 -1.90 0.00 -0.53 .130 0.77* -0.17 -0.13 0.02 -1.04 0.24 3.21 8.08 0 0.1 1.19 0.32 31 S48 CML144/CML312 0.02 -1.97 -1.09 -0.72 -0.57 .200 -0.29 -0.38 -0.10 -0.05 -5.16 -0.12 -1.48 2.69 0.15 0.66 0.22 32 S49 CML144/ZL130-23 -0.21 4.00* 4.86 -1.11 -7 .83-0.69 -0.09 0.12 -0.20 -0.09 -12.79 0.16 -1.91 91- 2. 0.06 -0.73 -0.09 33 S410 CML144/CML288 0.19 -0.69 0.47 -1.71 -1.980 .44- -0.26 0.48 -0.09 0.40** -7.49 0.13 -3.74 0.08- 0.44 -3.22 -0.38 34 S411 CML144/CML202 0.75 0.08 1.27 -1.33 5.38 7 0.2 -0.25 0.63 0.03 0.02 12.05 1.52** 3.79 -2.25 6-0 .0 0.96 -0.47 35 S56 CML161/CML172 0.17 0.04 -0.10 -0.72 -2.21 060 . 0.13 0.38 0.21 0.66** -11.01 0.43 -1.29 2.03 00 .1 1.09 -0.16 36 S57 CML161/CML448 0.06 -0.47 -0.29 -0.69 -6.42 .870 -0.47 -0.79* 0.24 -0.18 3.88 -0.02 2.33 -6.86 0.34- 0.19 -0.50 37 S58 CML161/CML312 0.00 0.84 2.53 -0.97 -10.68* .640 0.79* 0.00 0.26 0.25* -12.02 -0.55 -0.14 0.91 0.45- 1.32 -0.18 38 S59 CML161/CZL130-23 -0.08 -3.30* -3.42 -0.47 286 . -0.10 0.16 0.00 -0.05 -0.29* 9.57 0.73 -5.81* .38-2 0.63 1.16 0.09 39 S510 CML161/CML288 -0.25 5.09** 6.78 0.66 -6.60 0.17 -0.34 -0.07 -0.27 -0.30* 0.27 0.03 2.85 3.08 .56 0 -3.73 -0.28 40 S511 CML161/CML202 -0.18 0.61 -0.63 0.58 5.22 850 . -0.36 0.01 0.43 -0.18 1.06 -1.11* -2.34 -4.43 .26-0 -0.05 0.62 41 S67 CML172/CML448 -0.11 -0.61 0.19 -1.57 1.73 .03-0 0.14 0.12 0.52 -0.06 9.99 1.60** 1.22 -22.36* .700* 1.12 0.79 42 S68 CML172/CML312 -0.12 1.37 0.34 0.16 9.48 0.15 0.41 -0.34 0.22 -0.13 5.09 0.06 1.96 4.63 -0.42 25 0. 0.36 43 S69 CML172/CZL130-23 -0.01 -1.10 -2.27 0.99 5 .11-0.58 -0.22 0.91** 0.24 0.33* 7.68 -0.65 1.81 1 .46-0.34 1.09 0.39 44 S610 CML172/CML288 -0.09 1.29 1.59 1.11 13.56* 0.42- -0.22 -0.15 -0.32 -0.18 18.05 -0.35 -1.46 9- 5.3 0.59 -3.80 -0.49 45 S611 CML172/CML202 -0.18 0.81 0.85 -0.96 -6.96 .220 0.76* -0.58 0.05 -0.06 -9.82 0.01 -2.26 -15.8-20 .23 0.89 -0.09 46 S78 CML448/CML312 0.23 -2.47 -0.85 -2.14 -1.07 .770 -0.19 0.24 0.24 0.03 -3.02 -0.39 -1.94 -1.29 .35- 0 0.35 0.28 47 S79 CML448/ZLL130-23 -0.06 0.72 0.87 -0.64 -100* .1 -0.05 0.18 -0.26 -0.07 -0.01 -7.76 0.90* 1.17 610 . 0.23 1.19 -0.70 48 S710 CML448/CML208 -0.29 3.78* -0.60 4.15 9.01 .180 -0.32 -0.32 0.04 -0.02 9.60 -0.80 7.05 4.55 6 0.1 -2.70 -0.32 49 S711 CML448/CML202 0.02 1.63 1.66 0.74 7.83 0.90-0.34 0.75* -0.92* 0.11 7.73 -0.94* 0.40 5.92 -60 .1 -0.01 0.33 50 S89 CML312/ZL130-23 -0.19 -2.30 -3.65 1.41 6.6-40 .08 -0.06 0.53 -0.37 -0.08 6.34 -0.64 3.10 -0.840 .11 0.33 0.62 51 S810 CML312/CML288 -0.37 3.42* 1.88 0.54 -3.24 .210 0.45 -0.53 0.40 -0.09 -2.96 -0.34 -5.80* -3.420 .04 -2.56 -0.51 52 S811 CML312/CM202 -0.04 2.61 2.81 0.46 3.24 0.75-0.58 -0.46 -0.56 0.03 9.16 0.03 5.24* 0.11 0.72* 1.12 0.14 53 S910 ZL130-23/CML288 0.27 -1.05 -3.40 2.70 3.3-80 .65 -0.19 -0.53 0.09 -0.13 -0.70 -0.05 4.49 2.470 .12 -2.72 0.02 54 S911 ZL130-23/CML202 -0.59 1.47 2.86 -0.71 -5 .1-30.08 0.79* 0.04 -0.20 -0.01 -7.57 0.81 2.69 0.1-70 .20 0.96 -0.33 55 S112 CZL999601/CML539 -0.32 2.85 2.88 -0.75 4-2 .5-1.70 -0.09 -0.06 -0.27 0.15 1.26 0.28 2.01 - 7.6-60.04 0.62 0.15 214 Appendix 17 continue d 56 S212 CML481/CML539 -0.17 -0.83 0.85 -1.98 3.08 1.81- 0.11 0.10 0.14 -0.03 1.16 -0.69 0.96 -16.91 20 0. 1.00 -0.03 57 S312 CML359/CML539 0.56 1.83 0.35 0.15 -6.63 02-3 . -0.32 -0.30 -0.07 -0.12 -9.94 -0.02 6.75* 11.740 .18 -2.71 -0.12 58 S412 CML144/CML539 -0.21 1.62 1.13 0.31 4.74 64* .9 0.19 0.00 0.43 -0.04 17.72 -0.52 -1.23 -2.76 0 0.40.43 -0.04 59 S512 CML161/CML539 -0.07 1.87 0.30 1.73 9.75 43-0 . 0.61 -0.21 -0.50 0.26* 14.52 -0.95* 1.46 5.93 .37-0 0.09 0.26 60 S612 CML172/CML539 0.07 -0.60 -1.55 0.86 3.23 .38-1 0.23 0.21 -0.22 -0.12 -2.70 -0.84 0.09 -1.33 16 0. 0.02 -0.12 61 S712 CML448/CML539 -0.34 -1.11 1.26 -0.78 0.02 0.83- -0.37 0.28 -0.19 0.04 -13.48 -0.79 -8.67** 30 .1 0.23 0.12 0.04 62 S812 CML312/CML539 -0.07 -1.13 -1.93 0.95 6.10 0.69- -0.61 0.07 -0.49 -0.03 12.29 1.68** -1.25 8 .690.11 1.25 -0.03 63 S912 ZL130-23/CML539 0.30 -0.93 0.46 -0.56 -0 .964.92** -0.24 -0.18 0.86* -0.07 -0.12 -0.04 -4.941 0.-18 -0.30 1.09 -0.07 64 S1011 CML288/CML202 0.23 -1.81 -3.61 1.42 -4.68 0.07 -0.21 0.23 -0.26 -0.02 -2.88 0.61 -5.34* 3.1-70 .27 -2.93 -0.02 65 S1012 CML288/CML539 0.28 -0.21 -2.01 0.90 -9.15- 1.24 0.26 0.26 -0.03 -0.08 -7.09 1.76** 6.19* 7 .25-0.38 -2.80 -0.08 66 S1112 CML202/CML539 -0.17 -4.25 -0.93 -1.57 1-8 .8 1.53 0.34 0.42 0.31 -0.22 -18.52 0.80 -1.45 - 2.090.12 4.53 -0.22 **P ≤0.01; *P≤0.05; GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daEyHs) =, ear height (cm), EPP = ears per plant (#), LB = leaf blight disease (1-5), G=L gSr ay leaf spot disease (1-5), GT = grain tex(t1u-r5e) , MSV = maize streak virus disease (1-5), P pHl a=nt height (cm), RL = root lodging (#), SWT = 100 seed weight (g), SH = snhge lpliercentage, SL = stem lodging (#), VIG = vig Rours,t = rust disease (1-5), # = num.b er 215 Appendix 18 Estimated general combining abiliteyc tesf ffor 12 inbred lines for grain yield and agmroinc otraits for optimal soil conditions in 2011/12 Line Name GY AD ASI DS EH EPP GLS GT LB MSV PH RL WST SH% SL VIG 1 G1 CZL999601 1.31** -0.87** -0.45** -1.32** -2.0 1 -1.54 0.12* 0.02 0.07 -0.09 -1.57 0.51 1.94 3.13 .56 0 -0.76 2 G2 CML481 0.23 -0.83** -0.18 -1.02** -3.00 -1.54 -0.12* 0.00 0.37** -0.01 -7.53 0.88 -0.66 3.47 1 .19 -0.86 3 G3 CML359 0.12 -0.07 0.35* 0.28 1.36 -1.54 0.13* 0.04 -0.07 -0.03 4.32 -0.26 1.55 0.13 0.69 -0.75 4 G4 CML144 0.01 -0.67** -0.15 -0.82** 0.10 -1.50 .107** 0.03 -0.26* -0.08 -2.64 -1.04 1.14 3.63 0.13 -0.85 5 G5 CML161 -0.57** 0.73** 0.32* 1.05** 3.05 1.38 0.-15* 0.06 -0.02 0.02 4.04 -0.52 -3.01* -4.36* -60 .5 1.22 6 G6 CML172 0.25 -0.33** 0.22 -0.12 -2.58 -1.53 00 .1 0.00 0.18 -0.08 -0.76 1.41 0.28 2.10 -0.08 -0.75 7 G7 CML448 -0.51** 0.93** 0.22 1.15** -1.58 -1.58 -0.17** -0.03 0.35** 0.01 -4.60 -0.72 2.00 -0.94 .3-80 -0.43 8 G8 CML312 -0.40* -0.77** 0.08 -0.68 -4.39 1.06 .1-30* 0.20* 0.10 0.07 -0.29 1.53* -1.33 -1.38 0.09 .970 9 G9 ZL130-23 -0.31 1.63** -0.25 1.38** 3.76 4.76 ** -0.15* -0.18 -0.17 0.17* 2.80 -0.06 -2.88* -5.73 **-0.54 1.81* 10 G10 CML288 -0.02 0.13 -0.25 -0.12 -0.21 -1.54 200**. 0.03 -0.05 -0.03 1.17 0.13 1.87 3.12 -0.48 66-0 . 11 G11 CML202 -0.01 0.33** -0.15 0.18 6.70* 1.46 230*.* -0.02 -0.17 -0.01 7.65 -1.99* -1.52 -0.72 -90 .1 0.68 12 G12 CML539 -0.10 -0.23** 0.25 0.02 -1.22 2.11 .2-30** -0.15 -0.36 0.06 -2.59 0.13 0.63 -2.44 -0.440 .39 **P ≤0.01; *P≤0.05; GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daEyHs) =, ear height (cm), EPP = ears per plant (#), GLS = gray leaf spot disease (1G-5T) ,= grain texture (1-5), LB = leaf blight disea(1s-e5 ), MSV = maize streak virus disease (1-5), P pHl a=nt height (cm), RL = root lodging (#), SWT = 100 seed weight (g), SH = snhge lpliercentage, SL = stem lodging (#), VIG = vig #o u=r ,numbe.r 216 Appendix 19 Estimated specific combining abilfietyc tesf of 12 inbred lines for grain yield and agmroicn otraits at optimal soil conditions 2011/12 Cross Pedigree GY AD ASI DS EH EPP LB GLS GT MSV P H RL SWT SH% SL VIG 1 S12 CZL999601/CML481 -0.90 -1.94** -0.53 -2.47** 4.06 1.60 -1.01** -0.12 -0.34 0.14 -8.35 -0.38 -02 .4 -6.36 -2.09 0.84 2 S13 CZL999601/CML359 0.61 1.30** -0.73 0.56 -8 .70 1.64 -0.08 1.13** 0.13 -0.01 -5.61 0.42 0.14 -0 .31 -1.92 0.72 3 S14 CZL999601/CML144 2.44** 2.90** 0.10 3.00** .-608 1.75 -0.89** 0.26 0.64** 0.04 -3.41 -1.46 5.30 -4.58 2.98* -0.17 4 S15 CZL999601/CML161 -0.18 1.50** 0.97* 2.46** 798. -1.27 0.21 -0.25 -0.23 -0.06 12.51 -0.65 5.17 64 0. -1.01 -1.07 5 S16 CZL999601/CML172 -1.16* -2.44** 0.07 -2.37** 6.84 1.60 0.01 0.83** -0.01 0.04 10.85 -0.25 -5.34 2.79 0.84 0.89 6 S17 CZL999601/CML448 -1.33* -2.70** 1.07* -1.64 ** 0.90 1.63 1.51** -0.24 0.03 -0.04 -0.78 -1.78 4-5 .1 -1.42 0.81 2.24 7 S18 CZL999601/CML312 -0.08 -1.67** -0.13 -1.80** 7.05 -0.93 -0.74* -0.10 -0.36 -0.10 16.08 -2.03 50 .5 5.01 1.01 -0.33 8 S19 CZL999601/CZL130-23 0.94 -1.40** -0.47 -1.*8 7* 1.84 -4.42 -0.48 -0.09 -0.16 -0.21 9.55 2.89 2 .85 6.78 1.31 -2.83 9 S110 CZL999601/CML288 -0.73 2.10** -0.47 1.63** .80 1.72 1.57** -0.60** 0.14 0.00 -11.18 4.70* -10 .3 -2.59 -0.42 1.47 10 S111 CZL999601/CML202 0.86 1.90** 0.43 2.33** 5-.421** -1.34 0.02 -0.64** 0.19 0.14 -23.20 0.15 13 .8 3.15 -0.37 -0.37 11 S23 CML481/CML359 -0.41 -0.07 -0.67 -0.74 8.83 .681 0.79* 0.03 0.14 0.07 10.25 -0.95 -3.56 1.66 4*6*. 4 1.66 12 S24 CML481/CML144 0.22 2.86** 0.50 3.36** 7.32 .617 1.48* -0.34 -0.35 -0.04 9.09 -3.50 -5.98 -7.77 1.01 1.27 13 S25 CML481/CML161 0.70 -2.54 0.70 -1.84 -2.83 .14-1 -0.09 -0.02 -0.05 0.36 0.37 1.99 -2.67 5.21 3 1.0 -0.47 14 S26 CML481/CML172 -1.35* -4.14** -0.20 -4.34 143. 1.64 0.71* -0.10 -0.16 -0.04 12.17 4.72* -8.63* -3.44 -1.79 1.82 15 S27 CML481/CML448 2.05** 3.26** -0.87 2.40 -160. 6 1.72 -1.46** 0.00 0.21 -0.13 -44.65** -3.81 2.51 -2.31 -0.82 -0.16 16 S28 CML481/CML312 -0.50 4.30** -0.40 3.90 -2.42 -0.96 1.29** -0.04 -0.51 0.31 6.87 8.60** -4.17 0 .0 1.38 -0.56 17 S29 CML481/CZL130-23 0.52 -0.10 1.60** 1.50 -71 .9 -4.56 -0.61 0.15 0.19 -0.29 11.78 -3.81 14.40** .116 -1.32 -2.73 18 S210 CML481/CML288 1.25* 0.06 -0.40 -0.34 3.23 .591 -0.89 -0.04 -0.01 -0.09 7.24 -0.33 8.26* 0.45 1.06- -0.09 19 S211 CML481/CML202 -0.76 -0.14 0.83 0.70 -5.22 1.20- -0.11 0.43* 0.54* -0.11 -9.54 -2.21 1.64 4.89 -1.67 -0.94 20 S34 CML359/CML144 -0.83 0.10 0.63 0.73 -2.54 91 .5 -0.25 -0.59** -0.21 -0.03 8.03 -1.03 -2.24 -2.53 -1.49 1.65 21 S35 CML359/CML161 -0.22 -0.30 0.50 0.20 0.11 27-1 . -0.33 -0.10 0.25 -0.13 -0.58 -2.21 9.31* -4.66 0.81- -1.25 22 S36 CML359/CML172 1.09* 1.10** 0.93 2.03 -9.66 .931 -0.86 -0.19 -0.36 -0.03 -15.58 -1.15 1.35 - 3.64 0.04 -0.29 23 S37 CML359/CML448 0.26 2.50** -1.73** 0.76 1.64 1.67 0.64 -0.25 0.01 0.39* 5.19 -1.35 1.26 3.82 1 0.0 0.72 24 S38 CML359/CML312 -0.59 -2.80** 1.07* -1.74 -42. 7 -0.76 1.06** 0.21 -0.05 -0.17 -4.98 4.40* -9.26* 4.00 -0.79 -0.68 25 S39 CML359/CZL130-23 -0.27 -0.20 -1.27* -1.47 .6102 -4.67 -0.51 -0.27 0.49 -0.11 9.26 -3.01 -0.07 .59 0 -0.82 -1.02 26 S310 CML359/CML288 0.33 -2.70** 0.73 -1.97 -131 .3 1.60 -0.46 -0.45** -0.38 0.10 -13.28 -1.20 0.25 1.6-2 0.44 0.62 217 27 S311 CML359/CML202 -0.11 -2.57** 0.30 -2.27 8 .52 -1.39 0.16 0.51** -0.16 -0.09 1.41 0.92 5.34 2.53 -1.17 -1.55 28 S45 CML144/CML161 -0.52 -1.70** -1.33** -3.04 850. -1.30 0.20 0.36* 0.43 -0.08 5.39 0.90 0.06 4.08 0.09 -0.81 29 S46 CML144/CML172 -1.07* -1.64** -0.23 -1.87 46. 7 1.72 -0.34 -0.55** -0.01 0.02 6.95 -2.36 1.58 04 .7 0.28 1.65 30 S47 CML144/CML448 -0.20 -1.57** -0.90* -2.47 -.1461 1.63 0.16 0.38* -0.15 0.44* -22.37 -1.23 -5.67 3.86 -0.76 0.83 31 S48 CML144/CML312 0.45 2.80** -0.77 2.03 -0.96 1.0-0 -0.25 -0.32* -0.21 -0.12 -6.02 -2.81 12.21** .081 -0.56 -0.57 32 S49 CML144/ZL130-23 -1.10* 2.40** -0.10 2.30 0-70 . -4.54 -0.15 -0.30 -0.33 -0.06 -6.78 6.10* 1.11 3.10- -0.59 -1.58 33 S410 CML144/CML288 0.03 -2.10** 1.23** -0.87 3-38. 1.66 -0.10 0.68** 0.29 -0.02 -0.84 3.25 -5.83 690 . 0.34 0.23 34 S411 CML144/CML202 0.99 -2.30** 0.13 -2.17 -7 .79 -1.29 -0.15 0.65** 0.18 -0.04 -1.62 3.70 3.98 0 .81 -0.61 -1.45 35 S56 CML161/CML172 2.14* 1.96** -0.37 1.60 5.61 1.3-5 -1.08** 0.10 0.45 -0.08 10.57 -1.88 9.38* 6 .29 -0.04 -2.25 36 S57 CML161/CML448 -1.36* -2.30** 1.97** -0.34 .-452 -1.31 1.92** 0.03 -0.01 -0.16 -2.52 4.25 -5.42 -4.43 -0.07 0.93 37 S58 CML161/CML312 -0.44 -0.60* -1.23** -1.84 2-51. -3.86 -0.83** 0.00 0.09 -0.05 -14.00 -3.66 -0 .23 1.10 -0.21 -1.97 38 S59 CML161/CZL130-23 0.34 0.33 -0.57 -0.24 -153 .1 -7.57 0.77* 0.51** -0.20 -0.33 -20.46 -0.75 0.13 11.11 0.76 -3.98* 39 S510 CML161/CML288 -1.22* 0.50 -0.57 -0.07 -130 .5 -1.28 -0.68* -0.34* -0.07 -0.12 -5.99 -1.60 -3 .66 0.90 0.03 0.16 40 S511 CML161/CML202 -1.43* 2.30** -0.33 1.96 281*.4* 25.24** 0.11 -0.37* -0.44 0.86** 23.01 -1.48 4-1.06** -29.65** 0.24 12.77** 41 S67 CML172/CML448 -0.41 -3.90** -0.93* -4.84 9-13. 1.75 0.22 -0.22 -0.29 0.11 4.35 6.65* -6.92 5- 6.5 -0.22 -0.11 42 S68 CML172/CML312 -0.73 -1.87** -0.13 -2.00 -8119*. -0.97 1.64** 0.75** 0.32 -0.12 -16.96 1.07 -41. 4 2.22 -0.69 0.66 43 S69 CML172/CZL130-23 2.09** 0.06 0.87 0.93 -7 .26 -4.62 -0.09 0.10 0.36 -0.23 -2.56 -1.35 5.22 4.61 0.94 -1.85 44 S610 CML172/CML288 -0.21 1.56** -0.47 1.10 -7 .46 1.68 -0.71* -0.42* -0.35 0.48* -11.92 -0.20 1.40 5.4-7 0.21 0.46 45 S611 CML172/CML202 0.18 3.36** 0.10 3.46 8.03 .3-61 1.07** -0.12 0.04 -0.04 6.46 0.92 -4.06 0.32 590 . 0.11 46 S78 CML448/CML312 0.17 2.86** 0.20 3.06 8.95 0-51 . -1.03** 0.01 0.19 -0.20 16.28 -1.46 4.72 -9.19 .610 -0.66 47 S79 CML448/ZLL130-23 0.52 -1.20** 0.87 -0.34 9174 . -4.61 -0.26 0.20 -0.27 0.02 20.28 -1.55 9.16* 17 7. -0.09 -3.00 48 S710 CML448/CML208 0.09 4.30** -0.47 3.83 13.74 1.74 -0.88** 0.35* 0.52 -0.10 13.65 -1.06 0.82 3 .01 0.51 -0.03 49 S711 CML448/CML202 -0.45 3.10** -1.23* 1.86 -14 .2 -1.26 -0.76* -0.35* 0.07 -0.13 2.34 0.39 -3.11 .4-82 -0.11 -0.20 50 S89 CML312/ZL130-23 -0.06 2.83** 0.00 2.83 -4 .48 19.17** -0.68* 0.00 0.50 0.63* -5.06 -1.13 -10.2 9* -17.23** -0.22 8.62** 51 S810 CML312/CML288 1.86** -3.00** 2.00** -1.00 02.73 -1.04 -0.63* -0.35 -0.12 0.09 18.97 -3.15 6 .13 4.93 -0.62 -2.09 52 S811 CML312/CM202 1.00 -0.20 -0.43 -0.64 3.30 .97-3 -0.51 -0.39* 0.18 -0.19 -5.81 -3.53 10.75** 08 .8 0.09 -2.44 53 S910 ZL130-23/CML288 -1.44** -0.40 -0.67 -1.07 .279 -4.65 1.97** 0.00 -0.33 -0.27 5.35 2.94 -8.83* -0.09 0.01 0.23 54 S911 ZL130-23/CML202 -0.55 -0.27 0.90 0.63 3.69 -7.60 -0.41 -0.37* -0.45 -0.29 3.30 1.72 3.81 7.35 0.06 -2.95 55 S112 CZL999601/CML539 -0.47 0.46 -0.30 0.16 3.52 -1.97 -0.12 -0.17 -0.01 0.07 3.54 -1.63 -3.63 0- 3.1 -1.12 -1.41 56 S212 CML481/CML539 -0.83** -1.57** -0.57 -2.14 8.-52 -2.05 -0.09 0.06 0.34 -0.18 4.77 -0.33 0.61 55 1. -1.09 -0.64 218 57 S312 CML359/CML539 0.15 3.66** 0.23 3.90 5.25 .0-22 -0.15 -0.02 0.14 0.01 5.88 5.14* -2.51 0.17 8 0.0 -0.59 58 S412 CML144/CML539 -0.41 -1.74** 0.73 -1.00 153 .9 -1.89 0.29 -0.22 -0.27 -0.11 11.58 -1.58 -4.52 75 2. -0.69 -1.07 59 S512 CML161/CML539 2.17** 0.86 0.27 1.13 -5.64 4.8-8 -0.20 0.10 -0.22 -0.21 -8.30 5.07* 2.00 9.39 0.01- -2.05 60 S612 CML172/CML539 -0.59 5.93** 0.37 6.30 6.75 2.0-3 -0.57* -0.15 0.00 -0.11 -4.33 -6.20* 10.44* .8-21 -0.16 -1.09 61 S712 CML448/CML539 0.65 -4.34** 2.03** -2.30 45. 3 -1.91 -0.07 0.11 -0.30 -0.19 8.24 0.94 7.80* 8.52 0.14 -0.57 62 S812 CML312/CML539 -1.07* -2.64** -0.17 -2.80 .4-24 -4.62 0.68* 0.25 -0.02 -0.09 -5.37 3.69 -5.99 0.71- 0.01 0.02 63 S912 ZL130-23/CML539 -0.99 -2.04** -1.17* -3.20 -15.45 28.07** 0.45 0.10 0.19 1.14** -24.66 -2.06 17.-49** -23.31** -0.02 11.09** 64 S1011 CML288/CML202 -0.55 -3.44** -0.10 -3.54 .8-57 -1.08 0.81* 0.95** 0.01 0.08 -3.50 -0.46 -9. 32* -1.25 0.33 -0.14 65 S1012 CML288/CML539 0.58 3.13** -0.83 2.30 -7 .22 -1.94 0.00 0.25 0.31 -0.15 1.50 -2.91 12.08** 1 .02 0.24 -0.84 66 S1112 CML202/CML539 1.08 -1.53** 0.67 -0.87 -7 .1 -0.09 0.29 -0.02 -0.63 -0.07 -2.10 2.88 1.30 4.94 3.45* -1.83 **P ≤0.01; *P≤0.05; GY = grain yield (kg h-a1), AD = days to anthesis (days), ASI = antheskisin-sgi linterval (days), DS = days to silking (daEyHs) =, ear height (cm), EPP = ears per plant (#), LB = leaf blight disease (1-5), G=L gSr ay leaf spot disease (1-5), GT = grain tex(t1u-r5e) , MSV = maize streak virus disease (1-5), P pHl a=nt height (cm), RL = root lodging (#), SWT = 100 seed weight (g), SH = snhge lpliercentage, SL = stem lodging (#), VIG = vig #o u=r ,numbe.r 219