Aspirin and ticlopidine are two commonly used drugs in the prevention of cerebral embolic ischemic events. No direct comparisons in a cross-over design of the effects of ticlopidine and aspirin on asymptomatic circulating cerebral microemboli are available. We investigated 53 patients with cerebrovascular disease. Twenty-six patients were dosed for 2 weeks, 300 mg aspirin once daily and then for 2 weeks, 250 mg ticlopidine twice daily. In 27 other patients the scheme was reversed. Transcranial Doppler monitoring (both middle cerebral arteries simultaneously for 1 h were performed at the end of the two weeks. The signal was recorded on digitalised audio tapes and analyzed blinded off-line. The number of embolic signals per hour and vessel was 15.7 under aspirin and 11.7 under ticlopidine (difference not significant). The correlation between the number of emboli under the two medications was high. The highest number of embolic signals was found in high grade carotid stenosis. In patients with a low number of embolic signals, reproducibility was low. A minimum of 7 embolic signals in one treatment group is required for further therapeutic drug trials to allow reasonable comparisons. This study may help to plan further therapeutic trials using emboli detection.