With the use of washed human platelets suspended in a buffered salt solution, we have studied the factors capable of acting directly at the platelet plasma membrane to produce aggregation as opposed to those which produce the platelet release reaction and secondary aggregation. Platelet aggregation was measured in an aggregometer and the release reaction followed by assays of the release products. Ca and fibrinogen will not independently induce aggregation in washed human platelets suspended in a buffered salt solution. However, in their absence ADP-, 5HT-and epinephrine-mediated aggregation will not proceed. Of particular interest was the finding that in the presence of Ca, threshold doses of thrombin will produce aggregation earlier if μg amounts of fibrinogen are added. Ca and fibrinogen therefore appear to be absolutely required for aggregation, whether the aggregation is produced by the addition or by the release of required components. If either ADP, 5HT or epinephrine is added singly to buffer-suspended platelets in the presence of Ca and fibrinogen, 10–20% aggregation is observed. The combination of either ADP-5HT, ADP-epinephrine or 5HT-epinephrine produces 60–70% aggregation. The addition of ADP, 5HT and epinephrine together produces 80–90% aggregation. No release reaction is produced by the addition of ADP or 5HT alone or in combination. A new finding in this study therefore is that 5HT and ADP can act directly at the platelet plasma membrane to produce extensive aggregation in the absence of any release reaction. Epinephrine does produce a small release reaction, and may therefore produce its enhancement of aggregation either by acting directly at the platelet plasma membrane or by the release of something other than Ca, fibrinogen, ADP, 5HT, or epinephrine. In this study using washed human platelets suspended in a buffered salt solution, no evidence was found for the requirement of a plasma protein other than fibrinogen in platelet aggregation.