Recent Submissions

  • Methods for the Identification of Important Areas of Plant Diversity in Ireland

    Walsh, Aidan; Teagasc Walsh Fellowship Programme (2016-06)
    The loss of biodiversity is an on-going global issue that not only results in the extinction of species but also threatens the ecosystem services and goods on which humanity depends. The global community has responded with ambitious targets to halt and reverse the loss of biodiversity. The causes of biodiversity loss are well understood and conservation measures can be deployed to protect biodiversity. However, conservation resources are limited and to be effective must be targeted to the most important areas of biodiversity. Using vascular plant distribution records for Ireland this research examined the distribution of plant species of conservation concern to determine if additional conservation measures should be targeted to these species. In an attempt to aid the targeting of conservation measures to the most important areas of biodiversity this research also investigated methods for the identification of priority areas for plant conservation. In an initial study vascular plant distribution records were collated and mapped for the island of Ireland. The tetrad scale (2km x 2km) data provided incomplete coverage for Ireland. Records for plant species of conservation concern were extracted and mapped at the tetrad scale for Ireland. The coincidence of the locations of the species of conservation concern and the distribution of areas designated for the protection of biodiversity in Ireland was examined. Between 22 – 40% of the locations of these species were found to occur outside of designated areas and for some individual species all of the locations occurred outside designated areas. The results indicated the importance of both designated areas and the wider countryside for biodiversity conservation. In particular the presence of species of conservation concern in non-designated areas highlights the need for conservation measures outside of designated areas. Subsequent research investigated a method for the identification of the most important areas of plant diversity at the tetrad and hectad (10km x 10km) scales in Ireland. A criteria-based scoring method was developed to characterise the landscape in terms of conservation value and identify the important areas of plant diversity. The effect of each of the criteria on priority area identification and on the representation of species (that is the extent to which species occur within a set of sites) within the priority areas was examined. The outcomes of this research identified a combination of criteria that allowed the identification of priority areas of high conservation value that contained a high species representation level. A subsequent investigation examined an alternative method for the identification of priority areas for conservation. A complementarity-based method using linear programming was used to identify the minimum number of priority areas in which species representation was guaranteed. Additional linear programs were formulated to identify restricted numbers of priority areas in which species representation was maximised. This research also developed a means of incorporating the outputs of the scoring method into the linear programming method. A combination of the plant distribution data, spatial environmental data, and a logistic regression method was used to build models to predict the distribution of the plant species of conservation concern group (SCC). The associations between the environmental data and SCC occurrence were examined and the predictive performance of both models was investigated using plant records in County Waterford. To conclude, this research showed that the tetrad-scale plant distribution coverage is patchy for most of Ireland. However, even with incomplete data the research indicated that conservation measures could be targeted to sites outside of protected areas. The plant distribution data can be used to characterise the landscape in terms of conservation value and both the scoring method and linear programming method can be used to identify priority areas for conservation. The scoring method and linear programming method can be combined to efficiently identify priority areas of high conservation value and the research showed the potential use of species distribution modelling for conservation planning in Ireland.
  • Morphological and physiological responses of Fagus sylvatica and Quercus robur seedlings to light availability

    Sevillano, Ignacio; COFORD; Teagasc Walsh Fellowship Programme (2016-06)
    The need to develop forest management systems other than clearfelling has resulted in a requirement for improved understanding of the potential of continuous cover forestry (CCF). One suggested method for the conversion of forest stands into CCF systems and for bringing under-performing forests into productivity is thinning in conjunction with underplanting. This study was an attempt to provide information on species suitability for underplanting of two important trees in European forestry: pedunculate oak (Quercus robur L.) and European beech (Fagus sylvatica L.). To determine the morphological, physiological and growth responses of these two species to different light conditions, beech and oak seedlings previously grown at full light for two years were covered by shading nets that provide different shade levels (62%, 51% or 28% of full light) or continued to be exposed to full light. The different shade levels were intended to mimic a range of underplanted conditions and the process of acclimation to shade was studied to provide information on the ecology and adaptation of underplanted seedlings. In addition to the controlled-shade experiment another study to determine the physiological responses of beech natural regeneration to shade was conducted under natural light conditions (from open gaps to closed canopy). Both oak and beech displayed similar acclimation in response to shade for most of the traits investigated. At the plant level, seedling acclimation to shade included higher biomass allocation to above than below-ground parts and greater energy investment on height than diameter growth. At the leaf level, seedlings grown under shade reduced their leaf thickness and photosynthetic rates per unit area and increased their specific leaf area. This increase in specific leaf area seems to be one mechanism that allows seedlings to perform well under shade conditions. Another acclimation to low light conditions was to increase the efficiency of the photosystem II under shade. Photosynthetic rates were higher and leaves were retained for longer in seedlings grown at full light than under shade. Hence, this probably led to a greater growth in the full light than under shade. Despite this greater growth at full light, the results of this study suggest that beech and oak seedlings would be able to acclimate and perform well if underplanted below overstories that reduce the available light to as low as 28% of full light without having any significant adverse effect on the quality of the final crop.
  • Examining the physiological and genetic response of maize to low temperature conditions

    Di Fenza, Mauro; Teagasc Walsh Fellowship Programme (2013-01)
    Maize (Zea mays) is an emerging forage crop in Ireland, originating in warmer climates. Under Irish climate conditions establishment can be problematic due to low soil temperatures at early stages of establishment. Maize varieties with improved chilling tolerance have been developed and are on the market, but maize in Ireland is still established under plastic and further varietal improvements are required to make this crop more economically viable. To date, varieties are selected principally by phenotypic traits rather than genetic response. Investigation of the physiological and genetic response of maize towards cold/chilling stress at early developmental stage, in particular the response of developing maize roots to cold stress, can make a contribution towards the understanding of the molecular mechanisms conferring plant cold tolerance. The objectives of this study were aimed to create, at first, an experimental design to test the physiological response to low temperatures, under controlled environmental conditions, of various commercial maize cultivars adapted to grow in temperate climates. Responses to abiotic stresses such as cold involve changes in gene expression, therefore, once indentified the hybrids showing contrasting degrees of cold tolerance, these were profilied to examine gene expression and identify possible cold regulated genes. The physiological experiments on twelve maize varieties identified four genotypes with contrasting cold tolerance. Microarray analysis profiling these varieties was used to identify up and down regulated genes under cold/chilling conditions. The stress induced by the cold temperature in the genotypes Picker, PR39B29, Fergus and Codisco was reflected only on the expression profiles of the two varieties with superior cold tolerance, Picker and PR39B29. No significant changes in expression were observed in Fergus and Codisco in response to cold stress. The overall number of genes up and down regulated in the two cold tolerant varieties amounted to 69, which were, however, divided in a group of 39 genes in PR39B29 and 30 genes in Picker, as the two varieties exhibited two different trancriptomic patterns in which only four genes (RNA binding protein, pathogenesis-related protein 1 and two unknown proteins) were shared, although not all with the same degree of regulation. No cold regulated genes ware detected. The gene expression of the four-shared genes was further investigated with qRT-PCR in order to estimate the expression pattern over time. Five time points were used to analyse the expression trend of the genes. The gene expression was not maintained over the five time points, but it was subjected to fluctuation. However, with the exception of the RNA binding protein gene, the expression pattern was similar between the two varieties, indicating a common response to chilling stress.
  • The Impact of Chemically Amended Pig Slurry on Surface Runoff, Leachate and Greenhouse Gasses

    O'Flynn, Cornelius J. (2013-05)
    In Ireland, the pig industry is concentrated in a small number of counties. Pig farms typically have a high stocking rate. Therefore, the disposal of slurry in a cost-effective and environmentally-responsible way is a serious issue for farmers. Slurry is commonly applied to land, but this may not be possible if the land is at, or approaching, phosphorus (P) saturation. As pig farmers dispose of slurry in the vicinity of their properties, most of the nearby land is at P saturation, so alternative treatment methods need to be utilised (e.g. constructed wetlands, anaerobic digestion, filtration) or the slurry needs to be transported to another location. These alternatives are not currently financially viable in Ireland. Existing legislation (S.I. 610 of 2010) and recent changes in the implementation of legislation governing the timing and quantities of slurry that may be applied to land, means that pig farmers will no longer be able to exceed the maximum legal application rate to land (from January 2017). European policy aiming to intensify pig production will only accentuate this problem. If pig farmers are forced, in exceptional circumstances, to land apply slurry to unsuitable land, surface and subsurface losses of nutrients and suspended solids (SS) may occur. This could be potentially problematic if the land is located in a critical source area (CSA), an area that is highly likely to pollute receiving waters. In these circumstances, a possible novel solution is to chemically amend the pig slurry prior to landspreading. This would mean that pig farmers may, in exceptional circumstances, utilise the land in the vicinity of their farms for landspreading, without releasing excessive nutrients and SS into receiving waters. However, knowledge gaps exist concerning the type of amendments to be used, the characteristics of the soil on which they can be most effectively used, and their impact on incidental (short-term) and chronic (long-term) losses of nutrients, SS and greenhouse gas (GHG) to surface and subsurface water and the atmosphere. Therefore, the aims of this project were to: (1) identify the most appropriate chemical amendments, and their addition rates, to reduce P losses in runoff from pig slurry based on effectiveness, cost and feasibility; (2) investigate the impacts of these chemical amendments on nutrient losses in leachate, soil properties and GHG emissions; and (3) identify suitable soil types on which to landspread chemically-amended pig slurry. Laboratory bench-scale experiments were designed to identify the amendments which had the potential to reduce P in overland runoff and to quantify the stoichiometric rates at which to add them to the slurry. Based on effectiveness, cost and feasibility, the amendments identified were alum, which reduced dissolved reactive phosphorus (DRP) in overlying water by 86%, poly-aluminium chloride (PAC) (73%) and ferric chloride (FeCl3) (71%). Following these bench-scale experiments, rainfall simulation experiments were conducted to quantify the impact of chemical amendments to slurry on surface runoff losses at various time intervals from the time of application. Poly-aluminium chloride performed best in these experiments. For the first time, the effect of these amendments on GHG emissions, soil properties and leachate was also examined. Chemical amendment did not adversely affect GHG emissions, soil properties or leachate from pig slurry, but FeCl3 increased nitrous oxide (N2O) and carbon dioxide (CO2) losses. Finally, a 3-mo incubation experiment was conducted using a range of soil types to examine the effect of amendments on the long-term plant availability of P in soil and P solubility. Alum reduced more water extractable P than PAC, but also resulted in less plant available P. Considering cost, surface runoff and subsurface leachate losses, GHG emissions and impacts on soil chemistry, PAC was found to be the most suitable amendment with which to chemically amend pig slurry. There is the potential, in combination with existing programmes of measures, to employ chemical amendment as a measure to mitigate the environmental impact arising from the landspreading of pig slurry. This should be conducted in targeted areas of the CSA and should take into account soil type and its chemical properties. Before implementation, these tests must first be validated in long-term testing at field-scale over a wide variety of soil types, and include repeated application and incorporation. At present, there is no provision in legislation for chemical amendments to be used as a mitigation measure in the land application of pig slurry, but if they are to be utilised, a regulatory framework will need to be introduced by the relevant bodies.
  • Evaluation of Mehlich 3 as a Micronutrient Extractant on Irish Grassland Soils

    Brennan, Dennis D. (2002)
    The use of multinutrient extractants has been increasing in recent years, Mehlich 3 (M3) being one that has gained wide acceptance. The objective of this study was to see how M3 compared with methods currently used in Ireland for Cu, Zn, Mn and Fe extraction, and to investigate if it could be used to determine available Mo. Samples from eight mineral soil types, four of sandstone/shale and four of limestone origin and some organic soils were analysed for the micronutrients Cu, Fe, Mn, Mo and Zn using M3 and conventional extractants. Herbage samples were taken from specific mineral soils and analysed for the same range of elements. M3 results showed good correlation with EDTA and DTPA extractable Cu and Zn, and with easily reducible Mn, but poor correlation with DTPA extractable Mn and Fe. It was not possible to measure Mo in the M3 extract. Inclusion of soil properties and interacting elements in multiple regression models improved the coefficients of determination. Different relationships between extractants were displayed for mineral and organic soils. All extractants were equal in their ability to predict micronutrient content of herbage. Differences between sandstone/shale and limestone soils in relation to herbage micronutrient content were also found; the better relationships were generally found on the sandstone/shale. Results are generally in line with published data, but disagree with those of some studies. M3 is subject to the same shortcomings as existing extractants, but it’s versatility and range does offer an advantage.
  • Ryegrass organelle genomes: phylogenomics and sequence evaluation

    Diekmann, Kerstin; Teagasc Walsh Fellowship Programme (2010)
    Perennial ryegrass (Lolium perenne L.) is the most important forage grass of temperate regions of the world. The main objective in breeding perennial ryegrass cultivars is to increase its biomass. Chloroplasts and mitochondria are two organelles of the plant cell that are actively involved in biomass production. Chloroplasts derive from cyanobacteria and are the location of photosynthesis in plant cells. Mitochondria derive from α-proteobacteria and are involved in cell respiration. Due to their evolutionary history both organelles still contain their own genome which is in general maternally inherited. The interest in chloroplast genome sequences increased in recent years because they offer a useful option for plant genetic engineering. The risk of transgene escape via pollen flow is reduced while the expression of the transgene due to the high number of chloroplast genome copies is increased (in comparison to nuclear genome transformation). Mitochondrial genomes are of special interest because they are involved in cytoplasmic male sterility. Cytoplasmic male sterility is a very important trait in plant breeding programmes because it enables the cost efficient production of hybrid seed. Additionally, both organelle genomes can be used for molecular evolution or phylogenetic studies, as well as for population genetic approaches. Therefore the major aim of this thesis was to sequence the entire chloroplast and mitochondrial genomes of L. perenne to provide sequence information for chloroplast genetic engineering approaches, insights into the mitochondrial genome of a male fertile L. perenne cultivar and to gather knowledge about sequence variation in both genomes that can be used to design new markers for phylogenetic and population genetic studies.
  • The Effects of Seed Treatment, Sowing date, Cultivar and Harvest date on the Yield and Quality of Sugar Beet

    O'Donovan, Timothy M.; Irish Sugar Plc; Germain’s U.K. (2002)
    Trials were carried out at Lyons Estate Research Farm during 1998 on sugar beet (Beta Vulgaris). The commercially available seed treatment ‘Advantage’ was tested across 3 sowing dates, 4 commercially available cultivars and 4 harvest dates. Trials were carried out in a factorial arrangement and subject to standard statistical analysis. Data on emergence, ground cover, yield and quality was analysed. Percentage emergence of seed was satisfactory in all experiments. The benefit of ‘Advantage’ treatment during the emergence stage of growth was clearly seen in all trials. On average ‘Advantage’ seed emerged 2-3 days earlier than untreated seed and reached the target population (>75,000 plants/ha 4 days sooner than the control seed. During the growing season, ground cover was measured until complete cover was reached. In the trials it was less clear as to the benefits of using the treatment ‘Advantage’, as it was seldom statically different from control seed. As yields can be related to the radiation intercepted, it is not surprising that there were not large differences in yields. Delaying sowing date resulted in decreased yields of clean beet, sugar and extractable sugar. The two triploid varieties, Libra and Accord gave better yields of clean beet, sugar and extractable sugar than the two diploid cultivars Zulu and Celt. Delaying harvesting gave increased yields of clean beet, sugar and extractable sugar. The ‘Experimental’ seed treatment was not superior to the ‘Advantage’ treatment. In only one of the experiments was ‘Advantage’ better than the control treatment in respect of yield of clean beet, in the time of harvest trial. This was not reflected in yield of sugar or extractable sugar. In all other situations, there were no differences between ‘Advantage’ and the control treatments. There was no benefit from using ‘Advantage’ seed in the current experiments. However earlier sowing dates require investigation with pre-treated seed.
  • Characterisation of Miscanthus genetic resources: a combined analysis of plastid and nuclear microsatellites, nrDNA sequences, flow cytometry and morphology.

    De Cesare, Mariateresa; Teagasc Walsh Fellowship Programme; National Development Plan (2012)
    Miscanthus is a highly important forage and horticultural genus of perennial grasses (Poaceae) primarily native to South East Asia. Miscanthus is under intense global investigation as a biomass source for renewable energy production and several breeding initiatives are underway to develop new genotypes optimized for improved biomass and tolerance to a range of environmental stress conditions. A collection of 128 accessions belonging to the genus Miscanthus was established in Oak Park, Teagasc, Carlow, in 2008 and was investigated for morphological and molecular variation. Morphological traits were measured at the end of the second growing season and were compared with herbarium specimens of Miscanthus. Vegetative and inflorescence traits were scored and analysed using basic summary statistics, tests of normality and Principal Components Analysis (PCA). A large degree of morphological variation was recorded in the collections. The PCA of herbarium specimens was able to separate some species from others but there was also considerable overlap among species in the ordination, especially M. sacchariflorus, M. sinensis, M. condensatus and M. floridulus. These are known to be closely related and can interbreed. The PCA of the specimens from the Oak Park collection was less informative because of missing data due to lack of inflorescences (accessions did not flower). It was clear that morphology alone is often insufficient to distinguish taxa especially when inflorescence characters and ploidy information is lacking. The ploidy level of the accessions in the collection was evaluated through flow cytometry. The ploidy included di-, tri- and tetraploids. All individuals labelled as M. ×giganteus showed a triploid status, together with the newly bred M. sacchariflorus×M. sinensis hybrids. Most M. sinensis were diploids. Miscanthus sinensis Tea-62 was triploid and comparable to the value of the M. ×giganteus. A different situation was found for other non-diploid M. sinensis, in particular four M. sinensis ‘Goliath’ and the M. sinensis ‘Zebrinus’ Tea-33. In these the ratio measured by the flowcytometer was in between the values of the triploid M. giganteus and tetraploid M. sacchariflorus standards. The ‘Goliath-like’ hybrid is likely an autotriploid with three M. sinensis haploid sets, whereas M. ×giganteus is an allotriploid that is supposed to have two genomes from M. sinensis and one from M. sacchariflorus, which has a lower amount of DNA per haploid genome. DNA sequences of the internal transcribed spacer of the nrDNA were obtained for 76 genotypes in the collection and compared for polymorphism. The SNPs were particularly VI useful for differentiating M. sinensis, M. sacchariflorus and M. ×giganteus accessions and in combination with ploidy and morphology offer high potential for taxon identification. To gather more markers for population level diversity and differentiation studies, new microsatellite markers for both plastid and nuclear genomes were developed. For the development of plastid markers the chloroplast genome information of Saccharum officinarum was used. The nuclear SSRs (nSSRs) were developed from the sequences of 192 clones obtained from microsatellite enriched library. New primer pairs for the amplification of nineteen nuclear loci and six chloroplast loci were developed. Both chloroplast (cpSSR) and nSSR primers were used to characterise DNA variation, to help establish gene pools and to better understand hybridization and introgression. Huge genotypic variation was found within the genus, mostly in the species M. sinensis. The markers showed wide utility across a large number of Miscanthus species and also some closely related genera. The analysis of the cpSSRs showed a high number of different haplotypes but with a clear bias in allele composition between M. sinensis and the two species M. sacchariflorus and M. ×giganteus, thus confirming M. sacchariflorus as the maternal lineage of the hybrid M. xgiganteus. The nSSRs were found to be highly polymorphic across the collection and transferable to closely related genera such as Saccharum. The new markers were also used in UPGMA clustering and Bayesian structuring analysis to group individuals according to their similarity. Three major clusters of individuals were defined using the Bayesian STRUCTURE analysis with nuclear markers (nSSRs) and two with plastid markers (cpSSRs). In conclusion, the morphological, ploidy, sequence and microsatellite results highlighted the high level of diversity still unexplored in the genus and have clarified taxon identity of many accessions in the collection. A large set of new markers have been developed for the plant breeding and systematics community. The newly developed markers will be useful to further explore this diversity and to select useful traits for breeding of new and improved genotypes for biomass production.
  • Characteristaion of the initial generations of recombinant inbred lines in perennial ryegrass (lolium perenne L.) using molecular markers and cytogenetics

    Anhalt, Ulrike C.M.; Teagasc Walsh Fellowship Programme; National Development Plan (2009)
    In this study three early generations of a recombinant inbred line (RIL) breeding programme have been characterised by cytogenetic methods, metabolite profiling and a biomass quantitative trait loci (QTL) study. In situ hybridization was used to analyse recombination and the structure of the parental and the F1 genomes. A metabolite profiling study via GC/MS was conducted to characterise the metabolite activity of the parental inbred lines and the F1 hybrid. Total metabolites were compared across a growth season in a replicated field design. A genetic map was constructed with 75 nuclear DNA markers in the F2 generation, which consisted of 360 individuals. 63% of the markers in the F2 population did not fit the expected Mendelian ratios and showed significant (p<0.05) segregation distortion. Fresh weight, dry weight, and dry matter were analysed with an alpha lattice design throughout the greenhouse and field experiments. Additionally, leaf width was recorded in the greenhouse experiment. Heritablities ranged for the biomass related traits between 78 and 95%. This study provides an insight into the recombination of the parental and the F1 genotypes. Different levels of metabolite activities could be found among the two parental inbred lines across three harvest dates. The QTL study indicates the position of biomass QTL related traits. Major QTL with log of odds scores >3 were identified on linkage groups 2, 3, 4, 5 and 7. About 30% of overall variation could be explained. Increased biomass yield is still one of the most important traits in any Lolium perenne breeding programme. The present QTL study can be used for fine mapping of biomass yield related traits in L. perenne. In the long term biomass yield can be eventually monitored and predicted with marker assisted selection for some of the QTL identified in the present study.
  • The Characterisation of Genetic Diversity of a Collection of Perennial Ryegrass (Lolium perenne L.).

    McGrath, Sarah Katherine; Teagasc Walsh Fellowship Programme (2008)
    Perennial ryegrass (Lolium perenne L.) is a member of the Poaceae family, is native to Europe, the Near East and North Africa and is grown in all the temperate climate areas of the world as a forage and turf grass. Due to its persistence, palatability and nutritive value for ruminants, it is a principal component of pastures, and the most important forage species in Ireland. The primary aim of this thesis was to characterise the level of diversity in a large genetic resource collection of L. perenne germplasm held at Teagasc, Oak Park. Molecular markers, both chloroplast and nuclear SSRs, biochemical characters (water soluble carbohydrate, crude protein, and dry matter), and morphological characters (vegetative and flowering) were used to characterise this diversity, as well as population differentiation, and geographic patterns. Levels of diversity in all systems were found to be high in this collection. Primers to amplify microsatellite markers from the chloroplast genome of Lolium perenne were designed and optimized using de novo sequencing and in silico sequences. With one exception, each locus was polymorphic with a range from two to nine alleles in L. perenne. The newly developed primer pairs cross-amplified in different species of Lolium and in 50 other grass species representing nine grass subfamilies. These markers were then used to characterise chloroplast genetic diversity at allelic and haplotypic level in 104 accessions of Lolium perenne, other Lolium species, Festuca species and ´Festulolium cultivars. Furthermore, genetic relationships between the accessions and biogeographic distribution of haplotypes were investigated using a range of population genetic diversity measures and an Analysis of Molecular Variance (AMOVA). An extremely high number of 511 haplotypes was detected in 1,575 individuals possibly attributable to natural and anthropogenic migration. Much of the L. perenne European ecotype diversity (61%) could be attributed to within population variance. Plastid gene pools and maternal lineages for L. perenne could be clearly identified. Evidence was found showing a most likely migration route of L. perenne into Ireland from southern regions of Europe northwards. 5 Morphological variation of 13 vegetative and reproductive traits was characterised for 2,481 individuals from 50 L. perenne accessions, a mixture of Irish and European ecotypes and cultivars. Considerable levels of among and within population variation was found across traits. Principal component analysis and UPGMA dendrograms were able to separate ecotypes from cultivars. Cultivars generally had later dates of ear emergence, better spring and summer growth, longer rachis length and more spikelets per spike than ecotypes. Correlation and regression analysis were used to assess relationships between traits and strong positive relationships were seen between reproductive characters, i.e. rachis length with spikelets per spike, florets per spikelet and glume length. The strong relationship between rachis length and the other reproductive characters suggested that rachis length could be used as a predictor for reproductive performance. Later flowering was correlated with improved spring and summer growth. Water soluble carbohydrate (WSC; glucose and fructose determined by HPLC), crude protein (determined via LECO analysis), and dry matter contents were recorded for 1,320 individuals, pooled into 132 samples from 33 L. perenne ecotypes and cultivars at five different harvest time points across the 2004 growing season. While, in general, the cultivars had higher WSC contents than the ecotypes, individual ecotypes did show potential to be used in breeding programmes, as they showed higher values than all other accessions at particular cutting points. In correlation analyses, positive relationships were shown between dry matter and glucose both early and late in the growing season, and this was in agreement with the amount of leaves compared to stem at these times in the growing season. PCA analysis allowed the separation either between cultivars and ecotypes, or between tetraploid cultivars and the rest of the accessions at four out of five cutting points. In the ANOVA analysis, cutting point was the most significant factor influencing the variation in the traits. Eight nuclear SSR markers were used to characterize genetic diversity in 928 individuals from 40 diploid ecotypes and cultivars of L. perenne. High levels of genetic diversity (0.82, Nei’s gene diversity, over all accessions) and high numbers of alleles (22.25 average number of alleles per locus) was found. An average polymorphic information content (PIC) value of 0.81 across all loci was found. When deviations from Hardy-Weinberg equilibrium were tested, the majority of populations 6 had an excess of homozygotes. Very low levels of linkage disequilibrium were found between pairs of loci tested. AMOVA analysis and F statistics were used to test partitioning of variation, and most variation was found within populations (e.g. 31% for glume length in ecotypes). UPGMA, PCA and STRUCTURE analysis all gave similar patterns of relationships between populations, where relationships with high bootstrap support on the UPGMA dendrogram were also seen in the other analyses. The overall results of the thesis are discussed in the context of plant breeding programmes and natural population genetic variation. Strategies for incorporation of the results of the thesis (and the novel markers developed within) into plant breeding programmes are suggested.