To date, the normal range for self-recorded blood-pressure values has yet to be defined. In this paper possible definitions of the normal range of self-recorded blood pressure are discussed. In large epidemiologic studies self-recorded and casual blood-pressure values should be compared to each other, and in prospective interventional studies blood-pressure values for the initiation of therapy should be defined. Before the termination of these studies we recommend the start of an antihypertensive therapy above 140/90 mmHg. Although definite normal values for self-recorded blood pressure are still missing, a broader use of this method seems to be justified because of the well-known pitfalls of the causal clinical blood-pressure measurement.