Of the three studies which have been reported in this paper, the most thorough and therefore the most instructive was that made upon the Rockefeller Institute group of 24 individuals. The ten carriers discovered in this group were found to fall into three categories; namely, chronic, intermittent and transient carriers. It is, perhaps, a matter for surprise, in view of the weight of evidence in the literature, that half of the carriers should appear in the chronic group, being constantly affected for periods over 2 years and continuing to carry throughout this period what was, to all tests, the same strain of microorganism. It has been shown that no claim of relief from the carrier condition can be based on three consecutive negative swabs at weekly intervals since apparent spontaneous "cures," as evidenced by negative swabs, may last for 4(1/2) months and finally be terminated by the reappearance of the same strain as that carried before. The effect of coryza and pharyngitis on the persistence and degree of the meningococcal infection has been studied and, while the results are scanty, indications have been found that coryza, unassociated with any increase in numbers of the nasopharyngeal pathogens or streptococci, causes no change in the number of meningococci present in the throat. On the other hand, a streptococcal pharyngitis or any infection in which other throat pathogens increase greatly in number is usually associated with a marked diminution or actual disappearance, whether temporary or permanent, of the meningococci from the nasopharynx. This is in accordance with the work of Colebrook and Gordon. Of the 26 carrier strains which were isolated in these three groups of individuals, only eight could be identified with Gordon's four types which are isolated from the majority of cases of meningitis. It is considered as certain, however, that the other 18 strains are to be regarded as true meningococci. Not only do they show the same cultural characteristics and fermentative reactions as the typical strains, but serological tests, especially that of absorption, have revealed that they are allied to the two main types, I-III and II, and can be regarded as belonging to the broad serological Groups I and II which include these typical Gordon types. Moreover, atypical Type II* strains, identical with those isolated from the nasopharynx of carriers, have recently been found to be the cause of two cases of frank cerebrospinal fever. Only five of the 26 strains belong to Group I while the other 21 are members of Group II. This is interesting in view of the work of Scott who found that Group II strains predominate in carriers during interepidemic periods like the present. In periods of epidemics the carrier strains from both contacts and noncontacts in the epidemic zone are more often of Group I and even more constantly tend to be of the typical Gordon types rather than atypical forms. As has been pointed out in an earlier paper (35), the viability of these carrier strains when planted in defibrinated rabbit blood is low as compared to the typical and freshly isolated meningitis strains. The exact significance of this fact is not known. It has not been possible up to the present to do comparative virulence tests between spinal fluid and nasopharyngeal strains owing to the absence of a sufficiently susceptible animal.