Variability in degree results, especially in the proportion of good (i.e. first class and upper second class) degrees awarded, between subjects and institutions is well-documented. Relatively little work has examined variability within a single discipline. Analysis of degree results in geography in United Kingdom universities between 1973 and 1990 reveals persistent and systematic variations between departments in the proportion of good degrees awarded. Differences in entry qualification provide some explanation, but important questions remain. These questions have implications for various areas of current policy concern including the validity of value added performance indicators and the future of the external examiner system.