We have retrospectively compared the blood pressure load derived from 24 hour ambulatory blood pressure monitoring in patients with all clinic blood pressure readings elevated with those with only some elevated pressures to establish whether clinic readings alone are good predictors of blood pressure status. Fifty-seven patients attending a district general hospital hypertension clinic who were not on anti-hypertensive treatment were selected. Between two and six clinic readings were taken over a period of 1-6 months. Forty out of 57 patients had at least one clinic diastolic blood pressure reading of < 90 mmHg and, of these, 14 (35%) had a high blood pressure load and 26 (65%) had a normal blood pressure load. Patients with all diastolic blood pressure readings > 90 mmHg totalled 17 and of these 11 (65%) had high load and six (35%) had normal load. Patients with clinic diastolic blood pressure > 90 mmHg were significantly more likely to be truly hypertensive on the basis of blood pressure load than if one or more clinic readings was below 90 mmHg (P < 0.05). Diastolic pressures have some predictive power as to the blood pressure status defined by blood pressure load, but even consistently raised diastolic pressures do not necessarily indicate hypertension. Likewise one or more clinic diastolic blood pressure < 90 mmHg does not assuredly indicate normotension. Twenty-four hour ambulatory blood pressure monitoring may have an increasingly important role in the assessment of hypertension.