Autoimmune diseases result from a combination of genetic, immunologic, hormonal, and environmental factors. Infectious agents may induce the breakdown of immunological tolerance and the appearance of autoreactivity. However, the specific relationship between infection and autoimmunity is still unclear. One of the mechanisms responsible could be molecular mimicry between the infectious agent and self. The concept of molecular mimicry is a viable hypothesis in the investigation of the etiology, pathogenesis, treatment, and prevention of autoimmune disorders. Immune-mediated (type 1) diabetes in humans and in non-obese diabetic (NOD) mice is polygenic and characterized by autoimmune destruction of insulin-producing pancreatic beta cells in islets of Langerhans. In NOD mice, a T-helper 1 (Th1)-based autoimmune response arises spontaneously against glutamate decarboxylase (GAD) concurrently with the onset of insulitis. Subsequently. this Th1-type autoreactivity spreads intra- and intermolecularly to other beta cell autoantigens, suggesting that a Th1-type response is responsible for the progression of the disease, whereas Th2 responses when experimentally induced are protective. In humans, a homology between GAD and the P2-C protein of Coxsackie B make a cause-and-effect molecular mimicry an attractive hypothesis. Evidence to support the concept of molecular mimicry in diabetes is reviewed.