High fidelity measurements of left ventricular pressure were made at increasing pacing rates in 21 patients with hypertrophic cardiomyopathy and a control group of 11 patients investigated for chest pain who proved to have normal hearts. In both groups the fall in pressure during isovolumic relaxation from the point of min dp/dt approximated closely to a monoexponential, and could be described by a time constant and asymptote. The time constant shortened and the asymptote increased as heart rate rose in both groups. The time constant was longer and min dp/dt less in the cardiomyopathy group than controls at all heart rates. In the cardiomyopathy patients min dp/dt, but not the time constant, was related to systolic pressure. During pacing, eight cardiomyopathy patients developed metabolic evidence of myocardial ischaemia, but indices of relaxation did not differ between these eight and the other 13 either at basal heart rate or the highest pacing rate. In 10 cardiomyopathy patients measurements were repeated at comparable pacing rates after propranolol (0.2 mg/kg). Left ventricular end-diastolic pressure and indices of contractility decreased after the drug, but the time constant did not change. Eight patients received verapamil (20 mg) after which there were substantial reductions in systolic pressure and contractility. Min dp/dt decreased in proportion to systolic pressure, but the time constant was unchanged. At the highest pacing rate before drug administration three patients had abnormal lactate extraction which was corrected by either propranolol (one patient) or verapamil (two patients). Despite abolition of metabolic evidence of ischaemia, relaxation did not improve. It is concluded that abnormal isovolumic relaxation is common in patients with hypertrophic cardiomyopathy, but its severity correlates poorly with other features of the disease. Abnormal relaxation is not the result of ischaemia, and pressure derived indices of relaxation do not improve after the administration of propranolol or verapamil.