1. The effect of dipyridamole on platelet function was measured in twelve normal subjects given 150 or 200 mg tablets as single and multiple doses, and in six subjects given single doses of 25, 50 and 100 mg and multiple doses of 50 mg 8 hourly. 2. Platelet aggregation was measured in response to ADP and collagen. In the subjects given 150/200 mg, the platelets were assayed for content of cyclic AMP and for formation of thromboxane after addition of collagen. The responses to ADP and collagen and the cyclic AMP content were assessed in both the presence and absence of added PGE1. The pharmacokinetics of dipyridamole were studied in all subjects. 3. One hour after 150/200 mg single doses of dipyridamole there was significant inhibition of platelet aggregation in response to both collagen and ADP. There was no detectable effect on aggregation at other time points or with lower doses of dipyridamole. The addition of PGE1 to platelets prior to testing did not enhance the effect of dipyridamole on platelet aggregation. 4. In multiple doses, dipyridamole (150/200 mg twice daily for 11 days) had no detectable effect on platelet aggregation. 5. Dipyridamole did not have any effect on platelet cyclic AMP content, whether or not PGE1 was added prior to assay. 6. Dipyridamole did not affect platelet thromboxane formation. 7. Plasma dipyridamole concentrations were maximal 1-2 h after ingestion, at the same time that inhibition of platelet aggregation was detected. The concentrations declined in a biexponential fashion, with a terminal half life of 24.1 +/- 1.9 h (mean +/- s.e. mean). In six of the 17 subjects, the mean steady state plasma concentration was less than 75% of the value predicted from the single dose data.