In 1918 Fleischer reported that after transmission from one generation to the next, myotonic dystrophy has an earlier onset and is more severe. The hypothesis put forward by Penrose in 1948 that 'anticipation' is caused by bias of index case selection was based on theoretical arguments only and has not been supported by clinical observations. This hypothesis was tested in a clinical and genetic study of 14 families with myotonic dystrophy. Excluding index patients, an earlier onset in the child was found in 98% of 61 parent-child pairs. A greater mean difference in age of onset was found with transmission via the father than via the mother. The comparison of the severity of the disease between parents and children was difficult because of the variation in symptoms but, in general, the disease was more severe in the child than in the parent. The penetrance of the abnormal gene was nearly complete in the 14 families combined, leaving little room for the observation of 'complementary' parent-child pairs in the future. Also in retrospect no complementary parent-child pairs were found in the first generations. Fertility was severely reduced only in the very early onset patient group and this selective infertility cannot be responsible for the total amount of anticipation observed. It is therefore concluded that anticipation may be inherent in the transmission of myotonic dystrophy. This has important consequences for genetic counselling.