NONE of the theories of blood coagulation can explain some important anomalies without the introduction of hypothetical factors which have yet to be isolated. Their disadvantages are most apparent when an attempt is made to correlate: (a) the one- and two-stage method of prothrombin estimation; (b) the difference between the use of Russell viper venom plus lecithin and brain thromboplastin in the one-stage prothrombin estimation; (c) the apparent increase in prothrombin concentration in haemophilic blood after clotting; (d) an apparent auto-catalytic generation of thrombin after clotting has commenced.