There is abundant experimental evidence that both the M protein and the capsular gel of Group A streptococci play significant roles in pathogenicity (1-7). That they enhance virulence by interfering with surface phagocytosis is indicated by the observations described in the preceding paper (8).Despite the general agreement that both the capsules and the M protein of Group A streptococci are antiphagocytic (9, 10), their precise relationship to virulence has remained obscure. Numerous strains have been described, for example, which appear to produce abundant amounts of M protein and yet are relatively avirulent for mice (1,(11)(12)(13)(14). Others possessing large capsules have, likewise, been found to be lacking in virulence (3, 15), as illustrated by the glossy variant of the $23 strain referred to in the accompanying paper (8). Indeed, it has been shown that streptococci which produce both capsules and M substance also may lack pathogenicity for mice. The T14 strain, used in this and the previous study, is just such an organism (8).In the case of pneumococcal and Friediander's bacillus infections, the situation is less complex (16), inasmuch as these organisms are protected against leucocytes by a single antiphagocytic factor (i.e., a carbohydrate capsule) (9, 17, 18). As has been emphasized elsewhere, however, the quantitative aspects of capsule formation have a direct bearing upon the virulence of organisms such as Type III Pneumococcus (19,20). This fact, taken in conjunction with the knowledge that a double antiphagocytic mechanism is involved in Group A streptococcal infections, has prompted us to restudy the quantitative and combined effects of the M protein and the hyaluronate capsule upon phagocytosis. The results appear to clarify the relationship of both factors to virulence.

Four strains of Group A streptococci, possessing different degrees of virulence for both mice and rats, were tested for susceptibility to phagocytosis on glass slides, in glass roller tubes, and on the surfaces of freshly excised tissues and moistened filter paper. All of the tests were performed in the absence of serum to exclude the possible presence of opsonins. Only under conditions which allowed surface phagocytosis to take place was there a correlation between virulence and susceptibility to phagocytosis. A similar relationship between virulence and surface phagocytosis was also demonstrable in vivo during the early stages of experimental streptococcal peritonitis. Systematic study of the evolution of the peritonitis revealed that its outcome was determined by the phagocytic reaction which occurred in the first few hours of the infection.