The magnitude of the IgM response to a variety of antigens, induced on the in vivo administration of lipopolysaccharide (LPS) to mice, is approximately proportional to the magnitude of the background response observed in untreated animals. This striking correlation suggests that the administration of LPS alone cannot in general induce a specific response in the absence of a background response. Such is the case for the antigen rat erythrocytes against which no LPS-dependent or background response is found. The response to a marginally immunogenic dose of rat erythrocytes, however, can be considerably enhanced by the administration of LPS. These observations are expected on the hypothesis that both background and LPS-induced responses are due to ongoing antigen-dependent stimulation in normal mice. This hypothesis is further supported by evidence suggesting that the LPS-dependent anti-sheep erythrocyte response is due, at least in part, to a particular antigen present on degraded mouse erythrocytes.