A one-stage macroscopic test for platelet agglutination was used to study the effect of thrombin and thrombin-cation mixtures on washed platelets. Conclusions regarding platelet agglutination are as follows: (a) Canine, bovine, or human thrombin alone does not cause agglutination of canine or human platelets. (b) Thrombin with calcium or magnesium causes rapid platelet agglutination. Both calcium and magnesium are active at physiologic concentrations. Divalent manganese or cadmium ions can be substituted for calcium or magnesium. (c) The agglutination reaction is affected but little by the species of origin of thrombin or platelets, or by variations in ionic strength or pH over a broad range. (d) Temperature at which the reaction is carried out is critical; optimal temperature for the test is 28 degrees C. (e) Agglutination is inhibited by high ionic strength, by pH values outside the range 6.4-8.6, and by temperatures outside the range 25-28 degrees C. High concentrations of calcium have a specific inhibitory effect. (f) Platelet agglutination time is as sensitive an index of thrombin concentration as is the fibrinogen clotting time. A comparison is made between divalent cations which influence platelet agglutination induced by thrombin, TAg', and TAg. A similar comparison is made of cations influencing the action of thrombin on the "substrates," fibrinogen, TAMe, and platelets.