Insulin-dependent diabetes mellitus (IDDM) is caused by destruction of the insulin-secreting β-cells of the islets of Langerhans. It is proposed that the disease is caused by nongenetic, probably environmental, factors operating in a genetically susceptible host to initiate a destructive immune process. These environmental factors may operate over a limited period in early childhood to induce the immune process that destroys the islet β-cell. Thereafter, there is a long prodrome before the onset of clinical diabetes, during which clinical, immune, and metabolic changes can be detected. If these proposals are correct, epidemiological studies should focus on environmental events in early childhood that might induce, or accelerate, the disease process. Moreover, it should be possible to identify, from an early age, changes in the prediabetic period that persist to diagnosis and have predictive value. The variable age of presentation of IDDM may, therefore, reflect different rates of disease progression rather than different ages of exposure to the critical environmental events. Those patients in whom the disease process is slow could present with IDDM as adults, develop diabetes that does not require insulin treatment, or even fail to develop diabetes altogether. These proposed features, if confirmed, would have important implications for the prediction of IDDM and raise the possibility that modulation of the destructive immune process could prevent progression to clinical diabetes.