The in vivo inhibitory effect of aspirin on platelet cyclo-oxygenase is irreversible and lasts for the entire platelet life-span. Reappearance of cyclo-oxygenase activity in blood after aspirin has been proposed as a measure of the formation of new platelets and as an indirect indicator of platelet survival. A delay of 24--72 h in recovery, however, has been observed and it has been suggested that aspirin might also inhibit megakaryocyte cyclo-oxygenase. To test this possibility, aspirin (100 mg/kg) or saline were administered i.p. to rats made thrombocytopenic 2 h later (platelet count less than 5% of basal value) by a specific antiplatelet antiserum. Malondialdehyde (MDA) and thromboxane B2 (TxB2) production by platelets was measured by spectrophotometry and radioimmunoassay respectively, during the period of platelet count restoration. By 24 h after thrombocytopenia was induced, platelet count was about 15% of basal values in control and aspirin-treated rats. However, while in controls MDA and TxB2 production was restored to about 20% of basal values, in aspirin-treated rats less than 5% return of activity was detected. A marked difference between the two groups was still found 96 h after induction of thrombocytopenia, when platelet count restoration was similar. Since aspirin disappeared very rapidly from the circulation, the delay in recovery of cyclo-oxygenase activity supports the hypothesis of a megakaryocyte effect of this drug.