Over a period of twenty years, 704 hypertensives have been diagnosed in a London suburban general practice. This is a rate of 162 per 1000 for adults aged thirty years and over. In males the rate was 138 per 1000 and in females it was 187 per 1000. In 28 % the hypertension was mild, 58% moderate, and in 14% it was severe. During the period of observation specific hypotensive therapy was used rarely and therefore this survey may represent the natural history of high blood-pressure. The observed/expected (O/E) mortality-rate was 1·85 (3·01 in males and 1·62 in females), but most of this excessive mortality affected those hypertensives who were less than sixty years of age at first diagnosis. The O/E mortality-rates were inversely related to age. The initial height of the diastolic blood-pressure was a prognostic factor in the under-sixties, mortality increasing with the rising diastolic pressure. However, there were no increasing mortality-rates with rising pressure in the over-sixties. The conclusions are that specific hypotensive therapy is indicated for those hypertensives who are under sixty years, in males more than in females, and particularly in the younger hypertensives with high diastolic blood-pressure. There is no strong case for hypertensives first diagnosed over the age of sixty to be treated with specific hypotensives, and such therapy is less necessary in females than in males. Since half of all diagnosed hypertensives are over the age of sixty, a policy of selective non-treatment of high blood-pressure would result in considerable saving of medical manpower and financial resources.