In the wake of failed clinical trials of a formalin-inactivated RSV vaccine in the 1960's, the trial's sponsors, noting that the vaccinees who had experienced enhanced RSV disease had moderate amounts of vaccine-induced antibody, concluded erroneously that antibody to RSV was harmful. This hypothesis was further justified by the observation that, in unvaccinated patients, the most serious cases of RSV disease were seen in infants under 6 months of age, a time when maternal antibody is still present. For nearly 15 years this hypothesis, which had not been subjected to experimental verification due to the lack of an animal model of RSV infection, remained conventional wisdom. The development of the cotton rat model of RSV infection provided the opportunity to examine the nature of the immune response to RSV. In a crucial experiment involving the parabiotic linking of an immune to a non-immune cotton rat, we found that complete resistance to pulmonary RSV infection was conferred to the non-immune partner via whole blood. Subsequent blood fractionation experiments showed that the protective factor resided in the immune globulins, and that the cellular component of the blood neither conferred protection, nor enhanced the protective properties of the immune globulins. Quantitative studies using graded doses of immune globulins given to infant cotton rats prior to RSV challenge showed that serum neutralizing antibody titers in excess of 1:350 conferred complete resistance to pulmonary infection, and provided the rationale for the target titer to be achieved in human clinical trials. The cotton rat model correctly predicted not only that RSV immune globulin could prevent serious pulmonary disease in high-risk infants, but also the levels of serum antibody required to achieve protection.