In 106 plasma samples obtained from patients on heparin therapy, monitoring by 2 methods (activated partial thromboplastin time and thrombin clotting time--APTT and TT) was compared. All patients in whom APTT indicated markedly higher plasma heparin concentrations than the TT were critically ill (group B): their main diagnoses included severe infectious disease, severe liver disease and extensive myocardial infarction. Patients with lesser discrepancies between the results of APTT and TT did not suffer from such severe conditions (group A). Cardiac surgery without major postoperative problems, limited myocardial infarction and uncomplicated thromboembolism were the main diagnoses in this group. In group B, non-heparin related prolongation of APTT was thought to be the main factor responsible for the overestimation of plasma heparin concentrations by this test. We conclude that in patients with severe infectious disease, liver disease or extensive tissue necroses (i.e. myocardial infarction), APTT cannot be recommended for laboratory monitoring of heparin therapy.