A well-conducted twin study has the potential to evaluate the relative contribution of genetic and environmental factors to a given disease. Many studies are inconclusive because of problems with methodology. This article reviews the principles underlying the classical twin study and then discusses potential pitfalls. Twin studies in three diseases are evaluated, namely, multiple sclerosis, diabetes, osteoarthritis. Studies in multiple sclerosis are thought to be inconclusive. In type 1 diabetes, genetic factors are very important and according to the single study in females with osteoarthritis, genetic factors are also significant.