The effects of DDD (fully automatic) and VVI (ventricular demand) pacing modes on exercise tolerance, symptom diary cards, and Holter monitoring were investigated in a randomised double blind crossover study of 16 patients who had had DDD pacemakers implanted because of frequent syncope. Eight patients presented with sick sinus syndrome and, with one exception, retrograde atrioventricular conduction and eight age and sex matched patients presented with 2:1 or complete atrioventricular block. Maximal symptom limited exercise in those with atrioventricular block was significantly higher after one month of DDD pacing than after VVI pacing. In those with sick sinus syndrome, however, maximal effort tolerance was not significantly different for the two pacing modes. In all but one patient with sick sinus syndrome sinus rhythm developed during exercise in VVI pacing. For both VVI and DDD modes maximal atrial rates were significantly lower in those with sick sinus syndrome. Palpitation and general wellbeing were significantly improved during DDD pacing in the eight patients with sick sinus syndrome. Shortness of breath was improved by DDD pacing in the eight patients with atrioventricular block but not in those with sick sinus syndrome. Holter monitoring showed that sick sinus syndrome patients remained in paced rhythm, either DDD or VVI, for most of the 24 hour period. DDD pacing was better than VVI pacing in sick sinus syndrome with retrograde atrioventricular conduction. Despite their ability to show sinus rhythm and inhibit their pacemakers on exercise patients with sick sinus syndrome are just as likely to have symptomatic benefit from DDD pacing as patients with atrioventricular block.