The size and shape of wheat seeds are important characteristics that are monitored in breeding programs due to the impact they may have on the two largest economic factors determining the success of a cultivar, quality and yield. Despite this importance, there are mixed reports on the ability of plant breeders to accurately phenotype as unique a characteristic as 'shape' or to effectively implement selection procedures based on the morphology of wheat seeds. Furthermore, the complex physiological and genetic relationships between individual components of seed morphology complicate the matter of effectively improving yield or quality via seed shape. Due to recent developments in both phenotyping procedures and methods of genetic analysis, plant breeders can now understand seed shape at the genotypic level and translate that knowledge into meaningful improvements to wheat breeding programs. In this dissertation, a thorough review of the literature surrounding seed shape in hexaploid wheat is provided, as well as results of experiments comparing different phenotyping methods, clarification of QTL underlying seed shape in both adapted and 'exotic' mapping populations, as well as preliminary work in phenotypic characterization of unique genetic resources for future mapping work using association mapping. This work adds to the current body of scientific knowledge by improving upon seed phenotyping methods, suggesting differential modes of action of QTL on specific dimensions of seeds, validating previously reported seed shape QTL, and showing the impact of divergent selection for seed fill period on kernel morphology. It is hoped this work will facilitate the further improvement of not only wheat, but all plant organs of economic importance.