Germplasm exploration and phenotyping in Trifolium species for the improvement of agronomic traits and abiotic stress tolerance : A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy at Lincoln University

Trifolium is the most important pastoral legume genus for temperate agriculture. However, there has been little effort into characterising variation in Trifolium accessions. A series of studies utilising pedigree and genomic data were conducted to analyse variation in Trifolium accessions in the Margot Forde Germplasm Centre and in white clover breeding populations. Pedigree analysis experiments were designed to develop pedigree maps, calculate inbreeding and kinship coefficients, calculate the effective number of founders and identify influencing founders and ancestors for Trifolium repens (white clover), Trifolium pratense (red clover), Trifolium arvense, Trifolium ambiguum, Trifolium dubium, Trifolium hybridum, Trifolium medium, Trifolium subterraneum and Trifolium repens x Trifolium occidentale interspecific hybrids. One of the earliest parental white clover accessions identified was ‘North Canterbury Type I’ from 1941. The Type 1 phenotype influenced the population structure of all white clover accessions in the genebank. The relatedness and inbreeding coefficients revealed distinct germplasm pools formed across time that are of interest to pre-breeding efforts. One germplasm pool contained the majority of commercial white clover cultivars, of which they had similar phenotypes. Inbreeding in red clover remained stable in the last three decades, and a relationship between inbreeding and new introductions into the collection was found (r = 0.62). Founding accessions were classed into three phenotype groups; English Broad, English Giant Hybrid and Cotswald Broad. The first synthetic form was identified from parental accessions from English Broad and English Giant Hybrid. Within the minor Trifolium species, kinship levels remained below 8%, and no inbreeding was found. T. ambiguum and T. medium had the highest cumulative kinship across the decades. The Australian cultivar ‘Monaro’ had a strong influence over all of the studied accessions in T. ambiguum. Two genome-wide associations studies were designed to test for phenotype-genotype associations. The Mainstay panel had 242 white clover half-sib families, and the Genetic Gain panel had 80 white clover cultivars. The Mainstay panel had stolon density and growth score phenotype data. In contrast, the Genetic Gain panel had yield, clover content, seed yield, flowering duration, the peak number of flowers, leaf size, seed per head, leaf marking, cyanogenesis, normalised transpiration rate and the fraction of transpirable soil water phenotype data. Population structure was identified in the Mainstay panel (k=2), but no subpopulation divergence was present in the Genetic Gain panel. Both panels had rapid linkage disequilibrium decay. A marker-trait association was identified in the Genetic Gain panel for cyanogenesis. However, when the marker was tested against the marker dosage, it was not significant. The results of the genome-wide association studies emphasise the highly variable and heterozygous nature of white clover populations and cultivars.