Several strains of Streptococcus sanguis, Streptococcus mutans, Streptococcus mitis, Actinomyces viscosus, and Actinomyces naeslundii plus fresh isolates of Streptococcus salivarius were surveyed for their abilities to deplete lysozyme from human-whole-saliva supematant. Bacteria were incubated in saliva for 60 min at 37°C and then removed by centrifugation, and the recovered supernatant solutions were assayed for lysozyme activity by using whole cells of Micrococcus lysodeikticus as the substrate. Mean lysozyme depletions by bacterial strains varied over a wide (eightfold) range. The greatest mean depletion of lysozyme (60 to 70%) was observed with S. sanguis (biotype I), serotype b of S. mutans, and the fresh S. salivarius isolates. The lowest mean depletion was noted with S. mitis (15%) and biotype II S. sanguis (ca. 30%). The remaining species and strains exhibited an intermediate degree of depletion. In studies with S. sanguis 903, lysozyme was depleted by normal or heated (90°C, 30 min) bacteria and could be recovered from the organism. Furthermore, under appropriate conditions, lysozyme depletion by cells at 0 and 37°C was very similar. On the basis of these observations, we concluded that depletion was due to the adsorption of lysozyme by the organism. With S. sanguis 903, lysozyme adsorption depended on the concentration of bacteria, time of incubation, and the ionic strength of the medium. The extent of adsorption, however, was independent of pH's of 3.9 to 8.3. When a low concentration of S. sanguis 903 was used, lysozyme adsorption reached saturation (4 pLg of adsorbed lysozyme per 107 cells) at 20 jig of lysozyme