Although considerable progress has been made in perfecting the one-stage assay for factor VIII (antihemophilic factor), there remain variables that influence test results within laboratories as well as reproducibility between laboratories which have not been adequately evaluated. The purpose of this paper is to elucidate certain aspects of this assay that have not received adequate consideration and to describe the authors' assay method in order to provide a basis for comparison with results from other laboratories. It appears that variability results from: (1) differences in coagulability of different batches of substrate plasma obtained at different times from the same individual; (2) instability of some batches of stored substrate or standard plasmas; (3) variation in coagulability among vials of stored substrate or standard plasma from the same batch; (4) variation due to non-plasma reagents and instrumentation used to execute the test.