Rheumatoid arthritis remains the most important form of arthritis seen in rheumatological practice in the developed world. It presents some tantalizing epidemiological features. It is a relatively rare disease particularly in young life. It has a marked female excess in all populations studied. There are suggestions that it might be declining in incidence, though the reasons for this are unexplained. The geographical distribution of the disease is remarkably homogeneous. There is evidence of a genetic factor as witnessed by familial aggregation particularly by increased disease occurrence in identical twins; though the large majority of such twins remain disease discordant throughout life. Whilst undoubtedly one of the most important genetic factors in explaining disease occurrence is located in HLA class II, genes encoded in this region are neither necessary nor sufficient on their own for disease development. The female excess remains unexplained but it suggests a hormonal basis for disease development. Factors supporting this hypothesis include the observed protective effect of the oral contraceptive pill, the increased risk in women who are nulliparous and the increased susceptibility to disease during the first three months postpartum. This latter finding may be linked with breast feeding: women who have breast fed, particularly after their first pregnancy, have an increased risk for disease development. It is possible that this is explained by a massive increase in circulating prolactin levels.