Blood cells are crucial to human physiology, with functions in oxygen transport, infection control, and wound healing. Molecular mechanisms endogenous to blood cells have been implicated in the aetiologies of cancer, infection and inflammatory and immune disorders. The genetic determinants of blood cell function have not been comprehensively characterised, because it is too difficult to perform direct assays of cell function in large population samples. High-throughput flow cytometry can be used to measure functionally relevant phenotypes such as cell granulation, nucleic acid content, and cell size. Many of these phenotypes are important for the diagnosis of diseases such as sepsis, Szary disease, toxic granulation, and myelodysplastic syndromes, or correlate with assessments of cell morphology from blood smear images. Here, I report the results of my genome- wide association study of 63 previously genetically unstudied blood cell flow cytometry phenotypes. I have identified associated variants in loci containing genes coding for established drug targets with known roles in white cell function and immunity. I have colocalised the association signals with blood cell transcriptomic, blood proteomic, and disease risk, identifying possible causal roles for molecular mechanisms endogenous to white cells in the aetiology of a range of immune disorders, including atopic dermatitis, multiple sclerosis and celiac disease. My results have utility in drug design and therapeutic target selection, demonstrated by examples including the replication of the mechanism of action of Daclizumab, a treatment for multiple sclerosis, and evidence for the role of IL-18R1 in aetiology of celiac disease. Furthermore, mendelian randomisation analyses suggest a causal role for blood cell flow cytometry phenotypes in the aetiology of coronary artery disease, lung cancer, and asthma. In addition to my work on flow cytometry traits, I report a major contribution to the largest ever GWAS meta-analysis of routine clinical haematological phenotypes, including 563,085 individuals. I performed primary and conditional analyses, identifying parsimonious sets of independently associated variants. This is the largest genome-wide association study study of clinical haematological phenotypes to date and identifies 7,122 association signals.