Persuasive evidence has accumulated demonstrating a strong negative association between rheumatoid arthritis and schizophrenia at the population level. Explanations for this phenomenon have taken into consideration immunological, biochemical, and genetic factors. In this article, we examine these and other factors in closer detail. We then propose hypotheses at the molecular level that might account for the negative association between the two diseases. These hypotheses may provide clues for our colleagues in molecular biology as they search for candidate genes, "anti-genes," and molecular mechanisms relevant to schizophrenia.