Soil amendments and agricultural practices that are utilised to enhance crop production, can impact, either directly or indirectly, on soil microorganisms which are responsible for a whole range of processes that maintain soil health. The main aim of the present investigation was to determine the effects of these amendments and cultural practices on microbial diversity, activity and community structure. The potential impacts of these practices on the pathogenicity of soilborne pathogens such as F. oxysporum, were also evaluated. Trials conducted in soil microcosms investigated the effects of 10 soil amendments and 2 cover crops on microbial organisms. Qualitative analysis of microbial diversity, using Biolog Ecoâ¢Plates and PCR-DGGE, showed distinct differences in treated soils, compared to control treatments. Fertilizer amendments, herbicides, fungicides, cover crops and lime were all found to have a noteworthy impact on soil microbial diversity. A major factor influencing microbial diversity was a cover crop, viz. Nemat (Eruca sativa). Quantitative analysis of microbial activity using fluoroscein diacetate analysis (FDA) also showed distinct differences in treated soils compared to the controls. All treatments, except Roundup and both cover crop treatments increased microbial activity in soil compared to the control treatments. The highest microbial activity was observed in soil treated with the herbicide 2,4-D, and the lowest in the negative control treatment and soil treated with Roundup. A field study investigated the effects of Roundup, crop rotation and soil compaction on soil microbial populations. Qualitative analysis of maize fields treated with Roundup showed distinct differences between soils treated at various periods prior to analysis, and those not treated at any time. Qualitative analysis depicting microbial activity and biomass also showed distinct differences between Roundup soil and no-Roundup soil. Microbial activity seems to be immediately inhibited by the application of Roundup while microbial biomass was found to be negatively affected in the long-term. Qualitative analysis of soils subjected to rotation and compaction, also showed noteworthy differences in the microbial diversity of treated compared to non-treated soils. Both compaction and rotation was found to be a major factor influencing microbial biomass. Quantitative analysis of soils subjected to rotation and compaction, also showed distinct differences in microbial activity and biomass of the treated soils compared to the controls. Compacted soil had a lower microbial biomass and activity than non-compacted soil. Significant differences in sorghum seed germination and vegetative growth was found between microcosm soils compared to a vermiculite control treatment. Microcosm soil treated with rye grass, LAN, Roundup and mustard, significantly increased sorghum seedling germination and reduced the pathogenicity of F. oxysporum. Limed microcosm soil significantly reduced seedling germination while super phosphate (SP), benomyl, lime and mustard microcosm soil significantly improved the vegetative growth of sorghum seedlings by reducing the pathogenic effect of F. oxysporum. Results from the present study confirm the potential impact of soil amendments and cultural practices on soil microorganisms and the effects they may have towards soilborne diseases. This information may provide insights on improving soil health while helping to increase crop production by means of alternative cultural methods to control soilborne pathogens.

Breeding for drought and low N tolerant genotypes in Zimbabwe is an important intervention that will see the country curbing the food in-security problem. Both CIMMYT and DR&SS maize germplasm was used in this study that was conducted in Zimbabwe in the 2009/10 and 2010/11 seasons. Evaluations were done under optimum, drought and low N conditions. One of the objectives was to estimate combining ability and heterosis for grain yield and other agronomic traits of white maize inbred lines under stress and optimal environments. Line x tester analysis of 23 inbred lines identified RS61P, NAW5885 (from DR&SS) and CML444, CML539, CML442, CML537 and CML548 (from CIMMYT) as having desirable GCA effects under both drought and low N conditions. Additive and nonadditive gene effects were important in the expression of traits across all environments; however non-additive gene effect assumed a more important role in the expression of traits under stress environments. The single crosses RS61P/CML444 and 2N3d/CML548 were identified as potential testers for the SC and N3 heterotic groups respectively. In the analyses of G x E and stability using AMMI and GGE biplot the same single crosses were identified as the most stable. There were three mega environments identified within the test environments and Agricultural Research Trust farm site was the most powerful in discriminating genotypes. Genetic diversity amongst the 23 inbred lines was examined using 14 morphological traits and 1 129 SNP markers. The morphological data revealed variability amongst inbred lines that could be manipulated through selection and hybridisation. Variability was further substantiated using PCA where the overall diversity could not be explained by a few eigenvectors and the major contributors were grain yield, texture, ear aspect, common rust, GLS and anthesis days. Euclidean and Rogersâ dissimilarity matrices based on morphological and SNP data respectively clustered lines related by pedigree together in some cases. The SNP dendrogram had a high goodness of fit value (r=0.87) compared to the morphological dendrogram, which showed that it grouped the lines efficiently, although at times it was not in agreement with the known heterotic grouping that was previously established using testers. The assessment of correlation between genetic distances, F1 performance, heterosis and SCA revealed significant positive correlations and regressions between SCA, MPH, HPH and per se performance of hybrids. The HPH and MPH also showed significant positive association and linear regression along with high coefficient of determination with per se performance of hybrids, especially under drought conditions. Correlations of genetic distances with MPH and HPH were too low to be of predictive value. An average of 112.29% MPH and 76.40% HPH were realised across environments and this was an indication of the potential of these inbred lines for hybrid development. The segregating lines at F3 stage were testcrossed to group A (CML539/CML442) and B (CML444/CML395) testers and testcrosses containing lines derived from DR&SS lines K64r, RS61P, NAW5885, SC5522 and CIMMYT drought tolerant donors based on DTPWC9 were generally amongst the best performing testcrosses in early and late maturing trials. Three-way hybrid performance was predicted from 11 single cross hybrids and results showed that there was significant but weak correlation between the predicted and the observed grain yield means and this could be explained by epistatic and significant G x E interaction, which were not taken into account in the prediction equation. Three-way cross hybrids identified as having superior performance under drought and well-watered conditions included RA214P/CML538//RS61P, RS61P/ CML444//CML538 and RS61P/CML444//CML539. However, there is still need to evaluate these hybrids under low N conditions before they can be recommended for release.

Information on the nature and causes of G x E interactions in the rainfed parts of the South African sugar industry were lacking. The aim of this study was to systematically analyse, identify causes, and explore more comprehensive methods of analysing and understanding the G x E interactions of sugarcane, in order to optimize future MET networks. Data from plant breeding selection trials and post-release evaluation trials were systematically analysed using various statistical approaches combined with the use of soil and climatic data and crop models. Statistical methods based on multivariate methodologies such as GGE biplot, AMMI, and pattern analysis, were used to explore the effects of different environmental factors on sugarcane performance and agronomic traits. The age at harvest was the main factor causing different genotypic responses in the midlands region, which was unique in comparison with the coastal and hinterland regions that shared similar characteristics. In the midlands region, two testing sites were identified as being redundant and were recommended for removal from the trial network in favour of a testing site in a frost pocket. In the coastal/hinterland region, three sites were identified as being redundant. Along the coast, time of harvest influenced G x E interactions, with yields showing stronger correlations to stalk population in the early season and stalk weight in the late season. In all regions, site x ratoon interactions accounted for more variation in yield than genotype x ratoon interactions, suggesting that variation in ratoon performance is influenced more by site differences than genotype differences. The repeatable component of G x E interaction (genotype x site) was larger than the non-repeatable components (genotype x ratoon and genotype x site x ratoon) on the long cycle program on the coast, however, this was not the case on the short cycle. This suggests that more effort be placed on identifying more diverse test sites on the short cycle. In addition to providing direct recommendations for the industry selection programs, this study also illustrated novel methods of understanding sugarcane growth in different environments. The benefits of using a crop growth model to characterize sugarcane METs for water stress were illustrated throughout the study. The further use of the crop model to establish sugarcane growth phases also proved useful, and is likely to be more valuable when diverse datasets are analysed. Trials were characterized in terms of basic climatic and soil variables, which proved to be invaluable in understanding the causes of G x E interactions. The characterisation of the current sites ensures that future site selection will be more rigorous, as plant breeders will be more aware of the conditions to select for or against. The study showed that the integration of empirical and analytical statistical approaches was more valuable than using either approach in isolation, as is conventionally done in sugarcane. Additionally, these techniques were applied across many trial series and shown to produce repeatable results. The different strategies used to investigate sugarcane trait relations in this study have not been reported elsewhere, and future sugarcane studies dealing with similar traits (or other traits associated with lower level plant processes) may benefit from applying these methodologies. Furthermore, the integration of these multivariate methods with the largest ever simulation of sugarcane METs has opened new doors for the combined use of crop and statistical models in sugarcane research â an area not previously explored for this crop. The study illustrated novel methods of identifying factors responsible for sugarcane G x E interactions and introduced new ways of characterizing sugarcane METs through the use of crop growth models and supplementary environmental data.

Stripe rust, caused by the fungus Puccinia striiformis f. sp. tritici, is one of the most important diseases of bread wheat (Triticum aestivum). In recent years the disease has reached a global distribution, also causing significant crop losses in South Africa since 1996. The deployment of disease resistant cultivars is recognised as an effective approach to minimise crop losses without the higher input costs associated with the application fungicides. The need for new sources of durable resistance has become apparent with the regular changes in the pathogen population and the emergence of more virulent races. Quantitative trait loci (QTL) mapping is a powerful tool for unravelling the genetic components responsible for disease resistance. For the purpose of this study, two cultivars have been identified with complete adult plant resistance (APR) to stripe rust. Previously, the South African spring wheat Kariega has been studied and the Lr34/Yr18/Pm38 gene and two major QTL, QYr.sgi-2B and QYr.sgi-4A, were identified in a Kariega X Avocet S doubled haploid (DH) mapping population. Cappelle-Desprez is an old European winter wheat for which resistance genes/QTL have been postulated. The objective of this study was to dissect the APR for stripe rust in these cultivars using a QTL mapping approach. The Kariega QTL were further characterised by increasing the DNA marker density in a more targeted approach. Additional simple sequence repeat (SSR) markers were incorporated in the genetic map and expressed sequence tag (EST) markers were developed for screening by means of single-strand conformation polymorphism (SSCP) analysis. The conversion of EST and Diversity Arrays Technology (DArT) markers to sequence tagged site (STS) markers were investigated to allow for high-throughput screening. In addition, selected SSR and DArT-STS markers were screened in a large Kariega x Avocet S F2 mapping population to improve the genetic map resolution in the QTL intervals. The QYr.sgi-2B and QYr.sgi-4A intervals have been delimited to 6.1 and 16.2 cM respectively, in the F2 mapping population with the aid of recombinant mapping. Resistance displayed by Cappelle-Desprez was studied in a Palmiet x Yr16DH70 recombinant inbred line (RIL) mapping population. The breeding line Yr16DH70, a Cappelle-Desprez derivative, is a more suitable parent in a spring wheat background. SSR and DArT markers were typed in the population. QTL were identified, and up to 75.2% of the phenotypic variance could be accounted for. A major QTL, QYr.ufs-2A, explaining up to 53.2% of the phenotypic variance was identified on chromosome 2A. The presence of the Yr16 gene on chromosome 2D was confirmed (QYr.ufs-2D) and additional minor QTL were detected on chromosomes 5B (QYr.ufs-5B) and 6D (QYr.ufs-6D). A minor QTL, QYr.ufs-4B was derived from Palmiet. Stripe rust resistance QTL from Kariega and Cappelle-Desprez provide valuable sources of resistance in adapted, spring wheat backgrounds to South African breeders. Improved characterisation of the QTL has led to the identification of QTL-associated markers, allowing for more efficient selection in marker-assisted breeding schemes, and it also paves the way for map-based cloning.

Maize and sorghum are important crops produced in South Africa with 8 million- and 125 000 tonnes being produced respectively, annually. Ear and head rot fungi can negatively affect yield and grain quality of these crops and also produce mycotoxins which negatively impact on human and animal health. This study focussed on the occurrence of aflatoxin producing A. flavus and A. parasiticus and fumonisin producing F. verticillioides and F. proliferatum from commercial maize and sorghum grain. The natural occurrence of fumonisin producing Fusarium spp. and fumonisin contamination of maize was quantified in 29 maize production areas of South Africa over a three year period. Higher fungal biomass and fumonisin concentrations were associated with warmer production areas such as Northern Cape, North West and some areas of the Free State where the average temperatures ranged from 29°C to 32°C. In the cooler areas of Mpumalanga, KwaZulu-Natal, Gauteng and some areas of the eastern Free State, where mean maximum temperatures ranged from 24°C to 27°C, fungal biomass and fumonisin levels were absent or low. High fumonisin levels, in excess of 2 ppm recorded at 10 localities over the three year period, are of concern because of possible mycotoxicoses in animals and carcinogenic effects in humans. Corresponding high fungal biomass values may also indicate infection levels that may reduce yields and cause grain discoloration, physical breakdown of grain structure and reduction of grain nutritional value. Maize cultivars differed in susceptibility to colonisation by fumonisin producing Fusarium spp.. This necessitated a better understanding of the role of environment and the physiology of differential responses of cultivars in relation to infection by fumonisin producing Fusarium spp. at different localities in order to identify maize production areas with a potential high/low risk of fumonisin synthesis. The qRT-PCR method used in this study quantified the biomass of fumonisin producing Fusarium spp. (pathogenic and/or endophytic) in maize kernels more accurately than the plating out method and will replace the plating out method in future research. To determine the effect of aflatoxin producing Aspergillus spp. and fumonisin producing Fusarium spp. and their resultant mycotoxins on sorghum production in South Africa, sorghum grain samples were collected from five cultivars planted at 21 localities in South Africa from 2007-2009. HPLC and qPCR results indicate that Aspergillus spp. and Fusarium spp. and their mycotoxins do not pose a threat to sorghum production in South Africa. Data from Chapter 2, together with meteorological data, were used in the development of a provisional epidemiological model to predict the risk of maize kernel colonisation by Fusarium spp. and fumonisin contamination. Fusarium colonisation of grain and fumonisin levels were related to prevailing weather conditions during early post-flowering and grain development stages, respectively. Both colonisation and fumonisin production were significantly inversely correlated with mean maximum temperature and minimum relative humidity during the critical growth periods. Our models were consistent regarding time of fungal infection and fumonisin production in each respective season (2007-2009), although it did not give consistent prediction values over seasons, which indicates variation that is not accounted for by the selected two weather variables. This is an on-going study and the continuous incorporation of data into this model should improve predictive values over seasons. Since no cultivars/lines have been identified with resistance to Fusarium ear rot of maize in South Africa, the disease remains difficult to control. To date, no fungicides have been registered for the control of ear rots in South Africa and the potential of prophylactic fungicides, generally applied for the control of foliar diseases, to reduce Fusarium ear rot of maize and fumonisin synthesis was investigated. No significant differences between sprayed and control treatments on colonisation of grain by fumonisin producing Fusarium spp. or fumonisin contamination were recorded. This lack of efficacy may be attributed to the timing of fungicide applications relative to the plant growth stages critical to the infection of kernels by ear rot pathogens as determined from our epidemiological model. Further investigation into the possibility of using fungicides for the control of maize ear rots are being undertaken including time of application, application dosage and different fungicide regimes. Such applications must contribute to a reduction in maize leaf diseases as well as maize ear rots and their resultant mycotoxins. Only when fumonisin legislation and incentives are introduced into South Africa, will these fungicide spray programmes be economically justifiable. Robust, field-based models to predict fumonisin producing Fusarium ear rot in maize grain have been elusive due to the complexity of interactions between numerous abiotic and biotic disease factors. The findings in this study could contribute to an understanding of these complex interactions, thereby creating new management strategies to prevent or reduce the growth of F. verticillioides and F. proliferatum at field level as well as reducing contamination of grain with fumonisins.

The Bowman-Birk family of plant protease inhibitors are serine protease inhibitors that can simultaneously inhibit trypsin and/or chymotrypsin. Unique features include an unusually high thermal stability, tolerance towards low pH and resistance towards the action of proteases. Bowman-Birk protease inhibitors have been implicated in various biotic and abiotic stress conditions in plants. LRW222 was recently identified by means of SSH in Puccinia triticina-infected wheat and was found to be homologous to Wali5, a wound- and aluminium stress-induced Bowman-Birk type protease inhibitor. The aim of this study was to determine the expression levels and role of the LRW222 protein in Pucciniainfected wheat. This study identified and validated stable reference genes for qPCR gene expression analysis in rust-infected wheat. qPCR is the preferred method to quantify mRNA levels, but requires validated reference genes for data normalisation. As prescribed by the MIQE guidelines, the GeNorm-Plus algorithm was used to examine the expression stability of six candidate reference genes in resistant Avocet Yr1 wheat infected with P. triticina, P. striiformis and P. graminis f. sp. tritici respectively. These reference genes were used in the gene expression analysis of LRW222. Results indicated induced expression in both resistant and susceptible wheat, indicating its possible involvement in the early general defence response of wheat. Co-immunoprecipitation and mass spectrometry was used in an attempt to identify LRW222-interacting proteins to shed light on its specific role in plant defence. Protein identification was inconclusive, but results indicated that under natural conditions, LRW222 may exist as multimers in wheat.

Eastern Africa, where sorghum is a significant arable crop, is arid and areas are at risk of desertification. Consequently, many valuable landraces of sorghum are at risk for genetic erosion. Genetic diversity plays a vital role in the success of any breeding programme. This study aimed to investigate the genetic diversity and genetic relationships in germplasm accessions among east African countries using morpho-agronomical and simple sequence repeat (SSR) markers in order to (i) understand the extent of genetic variation in different countries, and (ii) quantify the genetic structure and how the diversity is distributed among and within countries. The extent of regional patterns of phenotypic diversity was assessed in 1013 accessions using 13 qualitative and five quantitative traits. A total of 1108 sorghum accessions from Sudan, Kenya, Uganda, Ethiopia, Eritrea, Rwanda, and Burundi were assessed using 39 SSR markers. Results on qualitative and quantitative traits data showed that there was a high level of morpho-agronomical diversity among accessions studied. Analysis of variance revealed highly significant differences between accessions pooled over countries and between countries. Based on SSR data, Sudan had the highest genetic diversity and Rwanda the lowest. Levels of genetic diversity differed significantly, with most of the diversity being partitioned more within than between countries. Results suggested that despite sorghumâs predominantly autogamous mating system, countries clustered totally separately with almost no integration and therefore emphasized the role of farmersâ practices in the preservation of landrace identity and the favouring of genetic diversity. The morpho-agronomical and molecular marker data showed a high level of variation among accessions, and indicated that sorghum populations studied were a mixture of a large number of different genotypes. Therefore, future germplasm collection should take all levels of variation into consideration.

During the present study it was found that herbicides, applied at the recommended dosage, has an inhibitory effect on fungi in vitro. However, when the effect of herbicides on soil microbial and bacteria activity was tested in experimental microcosms, it was found that soil with a high organic matter content had a buffer effect on soil microbes. Microbial activity increased after glyphosate treatment in rich soil microcosms, indicating tha soil microbes utilized glyphosate as a source of carbon and nutrients or that the toxicity of glyphosate to specific soil microflora and microfauna increased the available nutrient base for increased activity of resistant organisms. No significant differences were observed in terms of rhizosphere bacterial activity between glyphosate resistant (GR) maize and soybean in comparison to their conventional counterparts. Clay- and organic matter concentration in soil did however have a significant influence. Soil with a high clay- and organic matter content had higher microbial activity than soil with low clay- and organic matter content. Higher ergosterol levels were also isolated from high clay- and organic matter soil than low clay- and organic matter soil. Dry root biomass of GR maize and soybean kultivars did not differ significantly from their conventional counterparts. However, when glyphosate was applied to GR maize and soybean, dry root biomass decreased significantly compared to GR maize and soybean which received no glyphosate treatment. The same pattern was observed in terms of ergosterol concentrations, thus indicating that glyphosate had a physiological effect on the respective crop roots.

Maize and sorghum are important crops in South Africa, meeting the dietary needs of millions of people as well as the needs of the feed industry. Fungi within the Fusarium graminearum species complex (FGSC) have recently become the subject of importance locally. Maize kernels and sorghum grain were collected from the National Cultivar Trial over three and two seasons respectively. A total of 558 samples comprising of six maize and seven sorghum cultivars were collected from 34 and 22 localities, respectively. Although different conventional detection, isolation and identification methods were utilized, very low numbers of FGSC isolates could be obtained from the grain. Molecular identification was based on the translation elongation factor-1α (TEF1-α) and the ammonia ligase (URA) genes. F. boothii and F. graminearum s.s. were identified from maize kernels while F. acaciamearnsii, F. cortaderiae and F. meridionale where associated with sorghum. Results indicated host-specificity between members of the FGSC. Total genomic DNA was extracted from all milled kernel and grain samples and analysed for the presence or absence of FGSC DNA using TaqMan real-time PCR. F. graminearum s.l. DNA was detected and quantified in over 47% of the samples. These findings contradict previous reports that stated that this group of fungi is of minor importance. Reasons may be improved detection techniques as well as changes in agronomic practices, such as rotation of maize with wheat in South Africa, or to other shifts in environmental factors. DON and ZEA concentrations were quantified using CD-ELISA. In the absence of an ELISA test for NIV, a selected number of samples were used in LC-MS/MS based detection and DON, NIV and ZEA were quantified. Concentrations varied considerably and no correlation between the two techniques was found. Based on CD-ELISA, maize and sorghum contamination with DON and ZEA differed significantly between seasons, localities and in some instances cultivars. An analysis of variance across all three seasons showed significant crop by locality interactions. In maize, high levels of DON were detected in the 2007/08 season, in particular at Delmas, while during the same season ZEA contamination of maize was significantly higher at Bothaville. In sorghum, DON and ZEA contamination was highest during the 2008/09 season, although both toxins were detected in the preceding season. In both the 2007/08 and 2008/09 seasons, Cedara was the most conducive locality for grain contamination with both toxins. Trichothecene chemotyping was conducted using both simplex and multiplex PCR. All the F. boothii and F. graminearum s.s. isolates from maize were DON/15-acetyldeoxynivalenol (15-ADON) producers and F. acacia-mearnsii, F. cortaderiae and F. meridionale isolates from sorghum grain produced NIV. DON can act as a virulence factor in plant disease and is usually associated with greater pathogenicity on plants than NIV producers. On the other hand, NIV is believed to be more toxigenic to both humans and animals. Isolates of the 15- ADON chemotype are reportedly less toxic that those of the 3-ADON chemotype. The 3- ADON chemotype was not recorded in this study. However, more FGSC isolates need to be collected from South African cereal grains and evaluated for their mycotoxin potential to establish safety guidelines for end users of these products. The results of this study showed that levels of field infection of maize and sorghum grain by the FGSC are of significance. Future studies need to quantify the relationships between members of the FGSC and their tricthothecene and ZEA production in South African maize and sorghum production systems including the role of cultivar choice, weather, rotation, tillage and other practices with the aim of establishing intervention technologies.

In a citrus breeding programme, information on the breeding value of available parents and the heritability of specific characters can be an important aid to the breeder when planning controlled crosses for cultivar improvement. By quantifying the genetic variation in a population the breeder can study the relationships between the hybrids and parents and gain an understanding of how certain characteristics are inherited. Therefore a study was undertaken to investigate the genetic variation in the progenies of six mandarin families, where female parent Kiyomi tangor was crossed with male parents Dancy, Hansen, Rishon, Roma, Shani and Sunburst mandarins. Data were collected over a two year period for the fruit quality characteristics of rind colour, fruit size, fruit shape and fruit internal quality. A significant level of variation was found between the families for all the fruit characteristics studied. The within family variation was larger than the between family variation and indicated a high level of genetic variation within the families, while the within tree variation was found to be lower than the within family variation. Data collected over a two year period showed significant year and year x family variation for many of the fruit characteristics. The intraclass correlation coefficient, relevant to selection within the families, was found to be fairly low for all the fruit characteristics studied and indicated that the variation found within the families was only partly genetic and non-genetic factors contributed to the variation in the phenotype. Therefore only two years of testing will not be reliable and a mean performance over multiple years data is recommended for effective selection. All the families showed an improvement in the population for rind colour, fruit shape and the fruit internal quality while half of the families showed an improvement for fruit size. All the families showed heterosis with some individuals being superior to both parents. Families with male parents Hansen, Rishon, Shani and Sunburst had a rind colour superior to the other families, while the Kiyomi x Rishon family had larger fruit with a more round fruit shape and the Kiyomi x Roma family had a superior internal quality. This study revealed the male parent to have a large influence on the progeny and male parents with a deeper orange rind colour, larger fruit size, better fruit shape and superior internal quality resulted in families with a greater improvement in the population. Therefore only male parents of the highest quality should be used as parents in breeding programmes in order to increase the chance of obtaining a superior new cultivar.