Fifteen patients with chronic stable angina pectoris and a history of reduced exercise tolerance in cold weather (cold intolerance) underwent symptom limited treadmill exercise tests at 20 degrees C and 0 degrees C in a specially constructed cold chamber while taking no antianginal medication. Their mean time to onset of angina (5.8 v 4.2 min), to 1 mm ST depression (5.1 v 3.8 min), and to peak exercise (7.4 v 5.7 min) was significantly shorter on exercise at 0 degrees C than at 20 degrees C. The double product of heart rate and systolic blood pressure at each of these end points was the same in both exercise tests. Eight of these patients were treated with nifedipine 10 mg three times a day for two weeks and then with propranolol 40 mg three times a day for another two weeks. Repeat exercise testing was performed at the end of each two week treatment period. The mean time (SD) to peak exercise at the end of the nifedipine treatment period was 9.1 (2.0) min at 20 degrees C and 8.5 (2.3) min at 0 degrees C. The double product at peak exercise was the same for both exercise tests. At the end of the propranolol treatment period the mean time to peak exercise was significantly less at 0 degrees C (7.8 (2.6) min) than at 20 degrees C (8.9 (2.4) min). The double product at peak exercise was the same for both exercise tests but was significantly less than that on nifedipine. Cold intolerance was shown in patients with a positive history by symptom limited treadmill exercise testing at 0 degrees C. It persisted when they were treated with propranolol, albeit to a lesser extent, but not when they were treated with nifedipine.