The incubation of penicilloylated mouse erythrocytes with normal human peripheral leukocytes gives rise to plaque formation in an agar medium. While the phenomenon appeared superficially to be like Jerne hemolysin plaque formation, several distinct differences were noted. First, the reaction was not dependent upon the addition of exogenous complement. In addition, plating of enriched lymphocyte populations did not favor plaque formation. In contrast, enrichment of granulocytes favored plaque formation. Furthermore, the addition of protein (bovine serum albumin or human serum) enhanced the plaque formation. Finally, the plaques appeared in normal individuals presumably not sensitized to either mouse or penicillin antigens. On the other hand, certain similarities existed. There was an obligatory incubation time at physiologic temperatures, both systems depended upon the presence of viable cells, and both systems seemed to have an immunologically specific component. Because of the correlation of plaque formation with in vitro erythrophagocytosis, it is postulated that this reaction represented a phagocytosis of penicilloylated mouse erythrocytes directed in some manner by an opsonin.