A study of humoral and cellular immune responses to respiratory syncytial virus (RSV) was carried out in a research day care center over a period of 10 years. To analyze the cumulative effect of recurrent RSV infections, children were grouped by years of potential RSV exposure. RSV epidemics were documented each year from 1967 through 1979. Most epidemics occurred in January and February, but occasional RSV isolations were made in December or as late as May. Neutralizing antibodies to RSV were assayed in serum obtained in the fall and spring of each year. These data indicate that the majority of children became infected in each of the first 3 years of life. Cellular immunity, expressed as RSV-stimulated lymphocyte blastogenesis, was assayed in fall and spring blood samples during the latter 4 years of the study. The mean lymphocyte stimulation index (SI) was near unity before the first infection and rose gradually during year 1, 2, and 3 to a level of approximately 4.0 thereafter. To determine whether in vitro measures of immunity to RSV correlated with resistance to subsequent infection, the effect of antibody level on infection rates was examined. The infection rate for children with pre-existing titers of< 16 was significantly higher than those with titers of 16 or greater, but infections also occurred in children with high titers. The effect of pre-existing cell-mediated immunity, as expressed by antigen-stimulated lymphocyte blastogenesis, also was analyzed. More children with low stimulation indices (< 3.0) were infected but the difference between low and high reactors was not significant. When the combined effect of antibody titer and SI on infection rates was examined, protection correlated with serum antibody but no added effect could be related to CMI. The relationship of lymphocyte SI and clinical illness was examined by grouping subjects according to clinical illness syndromes associated with proven RSV infection. SI were statisticaly similar regardless of age or illness syndrome. Similar comparisons with pre-illness antibody titers revealed an inverse relation between height of titer and severity of clinical illness. Data derived from this prospective study of RSV infection do not support the hypothesis that cell-mediated immunity contributes significantly to resistance to infection or to disease pathogenesis. Because neither systemic humoral nor cell-mediated immunity appear to act directly in modifying RSV infection and disease expression, it is speculated that local immune mechanisms within the respiratory tract may be involved.