Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / A cultura do trigo apresenta grande importância econômica no Brasil, sendo que o país produz, anualmente, cerca de 6 milhões de toneladas. Contudo, essa produção é suficiente para atender a aproximadamente 50% da demanda. Essa situação faz com que o Brasil seja um dos maiores importadores deste cereal. O melhoramento genético da cultura tem grande importância na tentativa de aumentar a produção, a produtividade e a qualidade do trigo produzido. Além disso, aumentar a eficiência dos programas de melhoramento é essencial para reduzir os custos e o tempo necessários para o lançamento de novas cultivares. Neste sentido, soluções para a correta avaliação em etapas onde há baixa disponibilidade de sementes ou onde o número de linhagens a serem avaliadas é grande são necessárias. Desta forma, os objetivos deste trabalho foram: 1) avaliar o ganho genético para a cultura do trigo no Brasil, nos últimos 30 anos; 2) utilizar modelos aditivo-dominantes, em gerações F2 e F3, na identificação dos melhores genitores para caracteres de importância agronômica e 3) avaliação de linhagens homozigotas em ensaios multi-ambientes sem o uso de repetições. Para todas estas análises foram utilizados modelos mistos. Para a análise do ganho genético foram utilizados dados de 126 cultivares brasileiras de trigo, lançadas entre 1984 e 2014. Estas cultivares foram avaliadas em 187 ensaios, conduzidos em 25 locais, distribuídos na Região Sul do Brasil, entre os anos de 2002 e 2014. O ambiente foi responsável por mais de 70% da variância e os genótipos apresentaram comportamento similar entre os ambientes avaliados. O ganho genético obtido foi de 33,9 kg ha-1 ano-1,o que representa 1,28% ano-1. Além disso, os dados indicam que não há estagnação no ganho genético para a cultura do trigo no Brasil. A análise, via modelos aditivo-dominantes, de gerações heterozigotas (F2 e F3) indicou cultivares e linhagens que apresentam elevados efeitos aditivos, que são os principais efeitos quando o objetivo é o lançamento de cultivares a partir de linhagens homozigotas. Para o caractere rendimento de grãos, se destacaram as cultivares TBIO Seleto, Mirante, TBIO Mestre, Sinuelo e Ametista, além das linhagens UTFT 0932, UTFT 0908 e UTFT 0944. Na análise de adaptabilidade, estabilidade e produtividade, as linhagens UTFT 1110, UTFT 1608, UTFT 1620, UTFT 1025 e UTFT 1691 se destacaram e seriam selecionadas em cada um dos ambientes avaliados. Contudo, as linhagens UTFT 1634 e UTFT 1405 estiveram entre as linhagens selecionadas no conjunto de locais, mas poderiam ter sido eliminadas caso o ensaio tivesse sido conduzido em um único local, com repetições. / Wheat crop has great economic importance in Brazil, producing annually about 6 million tons. However, this production is only sufficient to meet ~ 50% of demand. This condition makes Brazil one of the largest importers of this cereal worldwide. The genetic improvement of this crop has great importance in the attempt of increasing production, productivity and quality of wheat produced in Brazil. Furthermore, increasing the efficiency of breeding programs is essential to reduce costs and the time required to release new cultivars. In this sense, solutions are necessary for the correct evaluation in steps where limited seeds are available or where the number of lines to be evaluated is very hight. Thus, the objectives of this work were: 1) to evaluate the genetic gain of wheat crop in Brazil in the last 30 years; 2) to use additive-dominant models, in generations F2 and F3, to identify the best parents for agronomic traits, i.e., grain yield, hectoliter mass, thousand grain mass, plant height, among others; and 3) to evaluate homozygous lines in designs without repetitions in multi-environment trials. For all analyses, mixed models were used. Genetic gain was evaluated using 126 Brazilian wheat cultivars released 1984 and 2014. Cultivars were evaluated in 187 trials, conducted in 25 locations, distributed in the Southern Region of Brazil, between 2002 and 2014. Environment effects was responsible for more than 70% of the total variance and genotypes presented similar behavior in the evaluated environments. Genetic gain was of 33.9 kg ha-1 year-1, which represents 1.28% year-1. Moreover, results indicated absence of stagnation in the genetic gain in Brazil. Analysis of F2 and F3 generations with additive-dominant models show cultivars and lines with high additive effects, which are the main effects when the objective is to release homozygous cultivars. For grain yield, cultivars TBIO Seleto, Mirante, TBIO Mestre, Sinuelo and Ametista and lines UTFT 0932, UTFT 0908 and UTFT 0944 presented the highest additive effects. In the analysis of adaptability, stability and productivity, lines UTFT 1110, UTFT 1608, UTFT 1620, UTFT 1025 and UTFT 1691 would be selected in each of the evaluated environments. However, lines UTFT 1634 and UTFT 1405 were among the selected lineages in the set of locations but could have been eliminated if the trial had been conducted in a design with replications in a single location.

<p>Wheat is a major source of calories and protein for humans worldwide. Wheat is the most widely grown crop, with cultivation areas and production systems on every continent. The cultivated land area is vast because of its importance and adaptability to various environmental conditions. Global wheat production has not kept up with the growing population, provoking the need to develop new methods and techniques to increase genetic gains. The first research chapter of this Ph.D. dissertation involves performing genome-wide association studies (GWAS) to identify and examine transferability of marker-trait associations (MTAs) across environments. I evaluated yield and yield components traits among 270 soft red winter (SRW) wheat varieties. The population consists of experimental breeding lines adapted to the Midwestern and eastern United States and developed by public university breeding programs. Phenotypic data from a two-year field study and a 45K-SNP marker dataset were analyzed by FarmCPU model to identify MTAs for yield related traits. Grain yield was positively correlated with thousand kernel weight, biomass, and grain weight per spike while negatively correlated with days to heading and maturity. Sixty-one independent loci were identified for agronomic traits, including a region that with <i>–logP</i> of 16.35, which explained 18% of the variation in grain yield. Using 12 existing datasets from other states and seasons, in addition to my own data, I examined the transferability of significant MTAs for grain yield and days to heading across homogenous environments. For grain yield and days to heading, I only observed 6 out of 28 MTAs to hold up across homogenous environments. I concluded that not all marker-trait associations can be detected in other environments.</p><p>In the second research chapter of this Ph.D. dissertation, I dissected yield component traits under contrasting nitrogen environments by using field-based low-throughput phenotyping. I characterized grain yield formation and quality attributes in soft red winter wheat. Using a split-block design, I studied responses of 30 experimental lines, as sub-plot, to high nitrogen and low nitrogen environment, as main-plot, for two years. Differential N environments were imposed by the application, or lack thereof, of spring nitrogen application in a field, following a previous corn harvest. In this study, I measured agronomic traits, in-tissue nitrogen concentrations, nitrogen use efficiency, nitrogen harvest index and end-use quality traits on either all or subset of the germplasm. My data showed that biomass, number of spikes and total grain numbers per unit area were most sensitive to low nitrogen while kernel weight remained stable across environments. Significant genotype x N-environment interaction allowed me to select N-efficient germplasm, that can be used as founding parents for a potential breeding population specifically for low-N environments. I did this selection on the basis of superior agronomic traits and the presence of the desirable gluten quality alleles such as <i>Glu-A1b </i>(<i>2*</i>) and <i>Glu-D1d </i>(<i>5+10</i>).</p>

<p></p>Comprehensive information of breeding germplasm is a necessity to develop effective strategies for accelerated breeding. I characterized Purdue University soft red winter wheat breeding population that was subjectof intensive germplasm introduction and introgression from exotic germplasm. Using genotyping-by-sequences (GBS) approach, I developed ~15,000 single nucleotide polymorphisms (SNPs) and studied extent of linkage disequilibrium (LD)and hidden population structure in the population.The extent of LD and its decay varied among chromosomes with chromosomes 2B and 7D showing the most extended islands of high-LDandslow rates of decay. Four sub-populations, two with North American origin and two with Australian and Chinese origins, were identified. Genome-wide scans for signatures of selection using FSTand hapFLK identified 13 genomic regions under selection, of which six loci (<i>LT, Ppd-B1, Fr-A2, Vrn-A1, Vrn-B1, Vrn3</i>) were associated with environmental adaptation and two loci were associated with disease resistance genes (<i>Sr36 </i>and <i>Fhb1</i>).<br><p></p><div><br></div><div>The population was evaluated for agronomic performance in field conditions across two years in two locations. Genome-wide association studies identified major loci controlling yield and yield related traits. For days to heading and plant height, large effects loci were identified on chromosome 6A and 7B. For test weight, number of spikes per square meter, and number of kernels per square meter, large effect loci were identified on chromosomes 1A, 4B, and 5A, respectively. However, for grain yield<i> per se</i>, no major loci were detected. A combination of selection for other large effect loci for yield components and genomic prediction could be a promising approach for yield improvement.<br></div><div><br></div><div>In addition, the population was evaluated for FHB resistance under misted FHB nurseries inoculated with scabby corn across 2017-18 (Y1) and 2018-19 (Y2) seasons at Purdue Agronomy Farm, West Lafayette,in randomized incomplete block designs. Phenotypic data included disease incidence (INC), disease severity (SEV), <i>Fusarium</i> damaged kernels (FDK), FHB index (FHBdx), and deoxynivalenol concentration (DON). Twenty-five loci were identified at -<i>log</i>P ≥ 4.0 to be associated with five FHB-related traits. Of these 25, eighteen explained more than 1% of the phenotypic variations. A major QTL on chromosome 2Bi.e., Q2B.1 that explained 36% of variation in FDK was also associated with INC, FHBdx, and DON. The marker-trait associations that explained more than 5% phenotypic variation were identified on chromosomes 1A, 2B, 3B, 5A, 7A, 7B,and 7D. To investigate the applicability of other QTL with less signal intensity, the threshold criterion was lowered to -<i>log</i>P ≥ 3.0, which resulted in the identification of 67 unique regions for all traits. This study showed that the FHB-related traits have significant correlations with the number of favorable alleles at these loci, suggesting their utility in improving FHB resistance in this population by marker-assisted selection.The genotype and phenotype data produced in this study will be valuable to train genomic prediction models and study the optimal design of genomic selection training sets. This study laid foundation for the design and breeding decisions to increase the efficiency of pyramiding strategies and achieving transgressive segregation for economically important traits such as yield and FHB resistance.<br></div>

<p>Heterosis, or hybrid vigor, is a phenomenon observed in both plant and animal systems where hybrid offspring perform better when compared to their parents. For hybrid plants, this can result in increased biomass, crop yields, and vigor when compared to the inbred parents. Even though heterosis has been used in agriculture for over a century, the molecular mechanisms that result in hybrid vigor remain elusive even after years of investigation. A molecular understanding of heterosis is desirable because it will speed up the process of breeding compatible inbred lines for developing hybrid seeds, and it will provide us with the knowledge to potentially engineer inbred lines that can mimic the beneficial phenotypic effects of heterosis, eliminating the need for farmers to buy new hybrid seeds every year. The goal of this research project is to identify genes that are required for heterotic phenotypes in maize. Our working hypothesis is that a mutation in genes that are essential for heterosis will cause an altered heterotic phenotype in hybrid maize plants. To test this hypothesis, we applied combined approaches of EMS mutagenesis, trait phenotyping in field and controlled conditions, bulk segregant analysis, whole genome sequencing, and bioinformatics analysis. First, we applied a forward genetics approach to identify mutant hybrids with altered heterosis and detected potential causal genes <em>via</em> whole genome sequencing. We identified one mutation occurring in a protein coding gene (gene ID <em>Zm00001eb305590</em>) located in a region of interest on chromosome 7, whose genotypes across various samples assayed fit the observed segregation pattern of hybrid traits. This mutation leads to a moderate or high-level codon change, indicating that this gene may play a role in mediating heterosis in maize. By investigating this gene with further studies, the learned knowledge could speed up the process of hybrid maize breeding by selecting compatible inbred lines through sequencing or by engineering hybrids that have favorable alleles for this gene.</p>

<p>  </p> <p>Rapid climate change on a global scale is posing a considerable threat to forest biodiversity. Assessing physiological and genomic backgrounds of each tree is crucial for informing conservation and mitigation strategies to evaluate species or populations' vulnerability and adaptive capacity under climate change. The goal of my dissertation research was to use morpho-physiological and molecular approaches in combination with genomic background, as a backbone knowledge for enhancing the restoration and conservation of different hardwood tree species. The same approaches also led to a better understanding of mitigation strategies of tree species to evaluate their vulnerability and adaptability under climate change. To do so, the native <em>Juglans</em> species (<em>J. cinerea</em> and <em>J. nigra</em>), local species (Arizona walnut<em>; J. major, </em>California walnut; <em>J. hindsii</em>), exotic species (Persian walnut, <em>J. regia</em>) and its F1 interspecific hybrids were used as a case study to evaluate the level of cold hardiness in <em>Juglans</em>. Hybridization can integrate biotic and abiotic tolerance in plants and could be a potential forest restoration and conservation tool. Evidence from past studies in some F1 interspecific hybrids indicates that naturalized hybrids of Persian walnut with black walnut or butternut have higher level of tolerance to lower temperature than Persian walnut. The potential cold tolerance of native, local, exotic <em>Juglans</em> species and F1 interspecific hybrid using field, electrolyte leakage, qPCR, and genome analysis was investigated, and results presented in chapter 2. Differences in cold hardiness were observed in tested <em>Juglans</em> species, <em>J. regia</em> as an exotic species and <em>J. major</em> from Arizona maladapted in West Lafayette, Indiana. No sign of cold damage was observed in F1 interspecific hybrids or native species. Using morpho-physiological, molecular, and genome data, we confirmed that molecular and morpho-physiological data were highly correlated and thus can be used to characterize cold hardy trait in <em>Juglans</em> species. </p> <p>Although the native <em>Juglans</em> species are cold tolerant, with current trend of climate change and rapid tree migration to the northern range, it is not easy to predict how <em>Juglans</em> species may adapt to new environments and response to other biotic and abiotic stresses in future. A reference-genome assembly for nuclear and chloroplast genomes and cold hardy genes is presented in chapter 3. We used re-sequence genomes of 170 individuals collected from 20 <em>Juglans</em> species and <em>Carya</em> (as an outgroup) of the Juglandaceae family distributed in temperate-tropical forests of America and Asia. We integrate genome and temperature variables to identify a set of associated single-nucleotide polymorphisms (SNP), structural variations, and the geographical distribution of the variants in the genes related to local adaptation of <em>Juglans</em> across latitudes. Phylogeny analyses revealed that <em>Juglans</em> species were sorted based on their origin using the nuclear genome, cold-hardy genes, and organellar genome. <em>Juglans regia</em>, a native species of Asia and Europe, was distinct from other species and exhibited less genetic diversity than <em>Juglans</em> spp. of North America, based on whole genome and cold-hardy gene analysis. We identified the black walnut as a more diverse species and the California walnut and Persian walnut (<em>J. regia</em>) as less diverse species using selective sweep and heterozygosity analysis. Within <em>Juglans </em>species, those from colder areas exhibited higher diversity of cold hardy genes compared to the ones from warmer regions. Differences in genetic diversity among continents and latitudes did not follow a clear trend. Still, the level of gene diversity of <em>Juglans</em> from North America is higher than the species that originated in eastern Asia. We can use 65,000 nuclear SNPs variants in an ecological modeling system to predict genetic diversity and spatiotemporal shift of <em>Juglans</em> species in response to future climate change. These SNPs variants are helpful for forest tree breeding programs with aims such as marker-assisted selection (MAS), conservation or assisted migration in future.  </p> <p>Based on the findings of chapter 2 and 3, black walnut is the most diverse species with high genetic diversity in comparison with other <em>Juglans</em> species distributing across eastern forest of the USA. However, deeper knowledge of how this genetically diverse species will be affected by climate change is crucial. In chapter 3, we projected black walnut's current and future basal area. Utilizing machine learning, we tested different models using more than 1.4 million tree records from 10,162 Forest Inventory and Analysis (FIA) sample plots and 42 spatially explicit bioclimate and other environmental attributes. Ultimately, we used random forests (RF) model to estimate the basal area of black walnut under climate change. The mean of annual temperature and precipitation, potential evapotranspiration, topology, and human footprint were the most significant variables in prediction of basal area. Under two emission scenarios (Representative Concentration Pathway 4.5 and 8.5), the RF model projected that black walnut stocking will increase in the northern part of the current range in the USA by 2080, with a potential shift of species distribution range. However, uncertainty remains due to unpredictable events, including extreme abiotic (heat, drought) and biotic (pests, disease) occurrences. Our models can be adapted to other hardwood tree species to predict tree changes in the basal area based on future climate scenarios.  </p> <p>A similar approach of chapter 2, with a slightly different freeze test (whole plant freezing test) and use of cold-acclimated seedling was used in chapter 4. For cold acclimation, seedlings exposed to air temperatures progressively lowered for eight weeks (from 25.6/22.2 ºC to 8/4 ºC, day/night) and non-acclimated seedlings from sea level to 2,300 m, in tropical Hawaiʻi, USA to evaluate cold tolerance of koa. We also investigated gene expression using qPCR and wideseq sequencing in this study. Freezing tolerance varied significantly in non-acclimated versus cold-acclimated treatments across the elevation cline using the whole plant physiology-freezing test and gene expression. The level of freezing tolerance and the elevation at which seeds were collected were consistent with the frequency of freezing tolerance genes to facilitate variation interpretation in cold-hardy phenotypes. Findings of physiology and molecular data analysis for freezing tolerance of koa across the elevation gradient of the Hawaiian Islands provides insight into natural selection processes and will help to support forest restoration efforts. </p> <p>  </p>

<p>While fresh vegetables (FVs) consumption is essential for public health, some high-profile outbreaks that cause severe illnesses are related to their consumption. To illustrate, the Center for Disease Control (CDC) and Prevention has estimated 48 million cases of foodborne illnesses in the U.S. per year; of them, about 46% are associated with FVs. The economic impact of food safety issues, estimated at $51 billion annually, is due to medical costs, productivity losses, and loss of consumer trust (Hoffman et al., 2021). </p> <p>The proliferation of risk mitigation methods (GAP, HACCP), food safety policies (FSMA), and information (labels, media, government) out in the market today, suggests that the way consumers understand food safety might be different from what policymakers, researchers, and retailers try to communicate. In addition, consumers' heterogeneous perceptions and beliefs can make communication with policymakers, researchers, and industry stakeholders ineffective when assessing food safety risks. </p> <p>Given the high demand for FVs and the communication mismatch with consumers, it is crucial to understand how consumers value food safety when purchasing FVs. This study clustered FVs consumers based on their food safety concerns. First, a Principal Component Analysis (PCA) identified the most relevant food safety dimensions. Later, using the food safety dimensions, this study segmented FVs consumers based on their food safety concerns. Finally, through a Multinomial Probit model (MNP), this study provided the main factors driving cluster membership.</p> <p>Our results suggest the existence of four segments of FVs consumers: “Worriers” (45% of our sample), who highly valued all the food safety characteristics when buying FVs. “Labelers” (20.3% of our sample) mainly valued attributes related to nutritional and environmental characteristics. “Pretty Vegetables” (17.3% of our sample) searched for the best and safest produce possible. Lastly, “DIYers” (17.3% of our sample) valued the least variables related to convenience in FVs. Finally, some of the main drivers of cluster membership were related to demographics, consumption, information sources, and perceptions about food safety of FVs consumers. These results can help policymakers, researchers, and retailers communicate food safety information more efficiently among different segments of consumers.</p>

<p dir="ltr">Pedicels are the slender stalks that attach the fruit to the plant. They play a crucial role in fruit development. The characteristics of the pedicel comprise complex traits that are controlled by multiple genes. To study whether genetic control of fruit mass was via control of pedicel characteristics, we used two unique hybrid apple populations: ‘20 Ounce’ x ‘Prairie Fire’ and ‘Edward VII’ x ‘Prairie Fire’. Both ‘20 Ounce’ and ‘Edward VII’ x ‘Prairie Fire’ produce large fruit over 200 g, whereas ‘Prairie Fire’ is a small-fruited crabapple with fruit size less than 2 g. These populations offer the potential to investigate how pedicel attributes relate to apple fruit size. Previous work established a correlation between pedicel characteristics and apple fruit mass. Specifically, pedicel length showed an inverse relationship, while pedicel diameter was directly related to fruit mass. Shorter and broader pedicels were expected to offer the least resistance to water and nutrient flows. We hypothesize that among the genes that control fruit mass, some govern pedicel length and diameter. Quantitative Trait Loci (QTLs) analysis (Linkage mapping) was performed, and 5 QTLs were associated with characteristics such as fruit mass, pedicel length and pedicel diameter with LOD scores of 4 and above. That being said, no common region was associated with both the fruit mass and pedicel characteristics. This does not support our hypothesis and suggests that different regions may be controlling all these traits. Knowledge of QTLs and subsequently genes that affect fruit mass and pedicel characteristics in apple have potential applications in apple breeding and fruit production. The identification and manipulation of these genes holds the promise of developing new apple cultivars with improved pedicel traits and ultimately fruit mass and enhanced fruit quality.</p>

Abstract not available. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2009.

Cassava (Manihot esculenta Crantz ssp. esculenta) is the second most important food crop and a main source of income for the rural communities with potential for industrial use in the coastal region of Kenya. However, its productivity of 5 to 9 t ha-1 is low due to the low yield potential of the local cassava landraces caused by cassava brown streak disease (CBSD) among other biotic and abiotic constraints. Breeding for CBSD resistant varieties with farmer desired characteristics is hampered by limited information on the current status of the disease and farmers’ preferred characteristics of new CBSD resistant genotypes. In addition, there is a lack of an effective inoculation technique for cassava brown streak virus (CBSV) for screening genotypes for CBSD resistance. Information about the general combining ability (GCA) and specific combining ability (SCA) for CBSD above and below ground symptoms, fresh biomass yield (FBY) and fresh storage root yield (FSRY) (kg plant-1), harvest index (HI), dry matter % (DM %) and picrate score (PS) is limited and conflicting especially for the cassava germplasm in Kenya. These studies were carried out to update information on the status of CBSD, farmer’s preferences for cassava genotypes, and identify the most effective CBSV inoculation technique. In addition, the studies aimed to: determine the GCA and SCA for, and gene action controlling, the incidence and severity of above ground CBSD, root necrosis, FBY, FSRY, HI, DM %, and PS; and identify CBSD resistant progeny with farmers’ desired characteristics. A survey carried out in three major cassava-growing divisions in Kilifi, Kwale and Malindi Districts indicated that there was potential to increase production and productivity by increasing the area under cassava production and developing CBSD resistant genotypes that are early maturing, high yielding and sweet. In addition, CBSD was widely distributed, being present in 98.0% of the farms surveyed at a mean incidence of 61.2%. However, 99.0% of farmers interviewed lacked awareness and correct information about the disease. The genetic variability of cassava within the farms was low as the majority of farmers grew one or two landraces. Highly significant differences (P < 0.01) were observed among inoculation techniques for CBSV for which the highest infection rate of up to 92.0% was observed in plants inoculated by wedge grafting infected scion. Highly significant differences (P < 0.01) were observed among genotypes, between sites and their interaction for incidence of CBSD and root necrosis, while the differences among genotypes and the interaction between genotypes and the period of ratings were highly significant (P < 0.01) for the severity of CBSD and root necrosis. Above ground CBSD symptoms were not always associated with below ground CBSD symptoms and below ground CBSD symptoms were more severe at 12 months after planting (MAP) than at 6 MAP. Therefore, selecting cassava genotypes with resistance to below ground CBSD is more important than selection based on resistance to above ground CBSD and should be done after 12 months. Genotypes 5318/3 (exotic) followed by Msa140 and Plot4 (both local) had high resistance and can be used as new sources of resistance to root necrosis. Both GCA and SCA effects were highly significant with GCA sums of squares (SS) predominant over the SCA SS for most traits evaluated except for DM % at the clonal stage. These results indicate that although additive and non-additive genetic effects are involved in the inheritance of these traits, the additive genetic effects are more important except for DM %. Therefore breeding for CBSD-resistant genotypes that have characteristics desired by farmers in the coastal region of Kenya can be achieved through recurrent selection and gene pyramiding followed by participatory selection or use of a selection index that incorporates characteristics considered important by farmers. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2008.

Low soil nitrogen (N) and drought impede maize production in the small-scale farming sector in Zambia; and adoption of new cultivars with improved tolerance might enhance production. This study: a) assessed farmer preferences for maize cultivars; b) determined genotype x environment interaction effects among popular maize cultivars under contrasting soil fertility levels and; c) investigated landraces for tolerance to low N and drought using S1 selection. The study was carried out in Zambia from 2004-07. Farmer preference influencing the adoption of maize cultivars was investigated using both formal and informal surveys in Luangwa, Chibombo and Lufwanyama rural districts representing the three agro-ecological regions of Zambia. Focus group discussions and personal interviews were used to collect data on issues that affected maize production in these areas. It has been found that although farmers perceived landraces to be low yielding, they believed that they were superior to improved cultivars for: tolerance to drought; tolerance to low soil fertility; grain palatability; grain storability; and poundability. The need for food security, their inability to apply fertiliser, and their need for drought tolerant cultivars significantly (p ≤ 0.05) influenced farmers in adopting cultivars. The farmers would readily adopt cultivars that address these concerns. The predominant use of certain landraces (76%) reflected their superiority in meeting some of these needs. The performance of nine popular cultivars (three for each of hybrids, OPVs and landraces) under contrasting levels of soil fertility, across six environments (ENVs) in the three agro-eological regions, was evaluated. An ENV was defined as season x location combination. The fertilizer treatments were full fertilization, basal dressing, top dressing and nil fertilization. The cultivars exhibited significant non-crossover type of genotype x fertilisation interaction effects at three ENVs, while the genotype x fertilisation interaction effects, were non-significant at the other three ENVs. The cultivars exhibited dynamic stability by increasing grain yield (GY) when fertilization was increased. Landraces yielded higher than all open pollinated varieties and were generally higher yielding than two hybrids. Based on average rank for GY, the five highest yielding cultivars were MRI724, Gankata, MM603, Kazungula and Pandawe. Superiority of landraces revealed their genetic potential for GY under low soil fertility and they should be used as germplasm in developing cultivars targeting such environments. Ninety-six local landraces were selfed to generate S1 lines (2004/05 season) which were crossed to a tester (2005/06 season). Testcrosses were evaluated under optimal, low N, and drought conditions (2006/07 season). Data on GY, anthesis-silking interval, number of ears per plant, leaf senescence, leaf rolling, tassel size and grain texture were recorded in all the trials during the study period. Testcrosses, their S1 parents and landraces that were superior under low N, drought, optimal conditions and across environments were identified; these should be used to develop varieties targeted to a particular environment. Selection for tolerance to drought also selected for tolerance to low N. Selection for low N tolerance also selected for GY under drought and optimal conditions. Therefore, in selecting for tolerance to abiotic stresses, use of optimal and managed stress environments was effective. The following landraces were superior at 10% selection intensity: LR38, LR84 and LR86 (optimal, low N and drought conditions); LR11, LR35 and LR76 (low N and drought conditions); LR12 (optimal and drought conditions); LR40 and LR93 (low N conditions only); LR79 (drought conditions only) and; LR74 and LR85 (optimal conditions only). These landraces should be used as source germplasm targeting respective environments. Significant (p ≤ 0.05) positive general combining ability effects for GY under both low N and drought conditions were found implying that additive gene action conditioned GY under the abiotic stresses. The heritability for GY under low N (0.38), and drought (0.17) conditions, was low suggesting that selection based on GY alone was not effective. The genetic correlation for GY between optimal, and either low N (rG=0.458), or drought (rG = 0.03) environments, was low (rG < 0.5) suggesting that indirect selection would not be effective either. Therefore, use of secondary traits for selection is discussed. The study established that most farmers depended on local landraces for seed and would adopt low input improved varieties that yield higher than the landraces. Some landraces were found superior to some improved cultivars under contrasting fertilisation regimes. The study also found that landraces had genetic variation for tolerance to low N and drought. Landraces, S1 lines and testcrosses superior under low N, drought, optimal conditions and across environments were selected and they should be used to develop cultivars targeting respective environments. Policy implications of these results are discussed. / Thesis (Ph.D.)-University of KwaZulu-Natal, Pietermaritzburg, 2007.