We have previously shown that complement (C) activated plasma causes granulocyte (PMN) aggregation in vitro and that C5a is responsible. The C-induced aggregation of PMNs treated with cytochalasin-B (CB) is markedly enhanced and irreversible, and the magnitude of the response is proportional to the log (concentration of activated plasma), allowing use of this technique to detect C5a and hence C-activation. To compare the sensitivity of granulocyte aggregometry to that of more standard methods of detecting C-activation, we produced graded C- activation in vitro by treating fresh serum with varying amounts of zymosan. Aggregometry was the most sensitive index of C-activation, detecting C-activation, produced by 0.02 mg zymosan/ml of serum--1/10 that required to produce C-activation detectable by C3 immunoelectrophoresis (the next most sensitive technique). Granulocyte aggregometry may also be used to detect in vivo C-activation. We have found aggregating activity in plasmas from patients with systemic lupus erythematosus, immune vasculitis, transfusion reactions, and other conditions associated with in vivo C-activation, but not in the plasmas of normal subjects.