1. Immune sera produced in rabbits by treatment with our Culture Strain A of Treponema pallidum agglutinated not only the homologous strain, but also the Noguchi strains, and indicate a close group relationship of other non-pathogenic treponemata. Absorption experiments confirmed this, indicating a close relationship between the pallidum and the calligyrum. 2. Culture treponemata are not agglutinated to a much greater extent by the sera of syphilitic rabbits than they are by those of normal rabbits. 3. Culture treponemata are not agglutinated to any considerable extent by the sera of rabbits immunized with virulent treponemata. 4. The sera of syphilitic patients, especially those in the tertiary stages, agglutinate culture pallidum to a slightly greater extent than do those of normal individuals, but the culture pallidum is agglutinated to an almost equal degree by the sera of many individuals with diseases other than syphilis. We do not think that we could definitely distinguish the syphilitic from the non-syphilitic serum by the agglutination of the culture pallida, and therefore we do not believe that the reaction has any diagnostic value at present. 5. Immunization with culture pallidum, either general or local, does not seem to confer upon rabbits any considerable degree of resistance to inoculation with virulent treponemata. 6. Rabbits that have once exhibited lesions in the testis are not easily reinfected at the same site if reinoculation is practiced more than a month or so after apparent healing of the lesion. We believe that the experiments above recorded strongly suggest that resistance to syphilis in rabbits is a localized cell phenomenon not dependent upon a generalized reaction on the part of tissues remote from the site directly involved in reaction with the invading treponemata. Antibodies analogous to those formed in most bacterial infections appear in the general circulation in slight amount, if at all. The finding of many motionless treponemata in a few of the small lesions following reinoculation suggests the possibility of a purely localized formation of antibodies. This was expressed by Landsteiner some years ago when he spoke of the localized formation of agglutinins when they were absent in the general circulation. We hesitate to apply these results too generally to the conditions prevailing in human syphilis, but they contain the possibility of an explanation for the apparent skin immunity of the secondary period, and the later successive involvement of some organs and tissues when others remain normal and when external superinfection is successfully resisted.