There are marked geographical variations in rates of medical and surgical intervention at every level of aggregation and in every aspect of medical practice. These data raise a range of important theoretical, methodological, and policy issues. Much the same pattern of variation characterizes the prescription and consumption of therapeutic drugs. Data from a survey of general practice in New Zealand confirm the existence of extensive variability in prescribing. Multilevel techniques are deployed to isolate the specific interpractitioner element in this variability. Controlling for patient, diagnostic, and practitioner variables improves the predictive power of the model but does not reduce the extent of interpractitioner variability in prescribing rates. The existence of such variability raises questions about the role of clinical uncertainty and professional autonomy in the promotion of rational therapeutics in medical practice.