Experimental gingivitis studies support a pathogenic rôle for Actinomyces viscosus because of its numerical predominance in disease-associated plaques. The aims of the present investigation were to quantify specific crevicular IgG against A. viscosus before and after conventional gingivitis treatment and to determine whether salivary IgA and IgG against A. viscosus and A. naeslundii would be affected concomitantly. 6 subjects with generalized gingivitis were selected. Examinations were made before and after treatment and included collection of unstimulated saliva, paraffin-stimulated saliva and crevicular material as well as measurements of clinical parameters. The immunoglobulins were estimated by an ELISA assay using whole bacterial cells as antigen. Crevicular IgG against A. viscosus WVU 627 was demonstrated in pre- and post-treatment samples with a tendency towards increased values in the post-treatment samples. Salivary IgA and IgG against A. viscosus were also demonstrated in pre- and post-treatment samples. There were tendencies towards increased IgA values for paraffin-stimulated saliva and increased IgG values for unstimulated saliva in the post-treatment samples. Salivary IgA and IgG values against A. naeslundii ATCC 12104 were either not detected or barely detectable in both pre- and post-treatment samples. A. naeslundii B 74 IgG values were also rather low. As for A. viscosus WVU 627, anti-B 74 IgA values demonstrated a post-treatment increase for most subjects, especially for unstimulated saliva for which all post-treatment values were elevated. It is interesting to note that saliva samples showed changes similar to crevicular material, considering that salivary IgA is most likely derived from a different source. Together, elevated crevicular and salivary antibody concentrations might account for some of the beneficial effects of periodontal therapy by scaling.