Measurement of the in vitro anticoagulant activities of low molecular weight (LMW) heparins against an unfractionated heparin (UFH) standard has given problems of non-parallel assays, wide variability between laboratories and differences between methods. These problems have been largely overcome by establishment of a new LMW heparin standard against which other LMW heparin can be compared. The LMW heparin standard is calibrated with two separate potencies, one for anti-Xa and the other for thrombin inhibition assays. Animal studies have shown a correlation between the anti Xa and antithrombotic activities of several LMW heparins but, below a M.Wt. of approximately 4,000, this correlation breaks down. This may be partly due to the method of measurement of anti-Xa activity and partly due to the inability of very LMW fragments to potentiate thrombin inhibition, which has recently been shown to be an important component of antithrombotic action. However, substances with either anti-Xa activity alone or thrombin inhibitory capacity alone are less effective antithrombotic agents than heparin. Several studies in animals have shown that LMW heparins give less enhancement of bleeding than UFH, but this does not seem to be due to their anticoagulant effects and may be related to their relative effects on platelets.