Enzyme-linked immunosorbent assays for the detection of immunoglobulin M (IgM), IgA, IgG1, and IgG2 antibodies against bovine respiratory syncytial virus (BRSV) were used to measure antibody responses of calves after experimental or natural infection with BRSV. Serially collected sera, lung lavage samples, nasal and eye secretions, and feces were tested for the presence of these antibodies. Lung lavage fluids and nasal secretions were further examined for the presence of virus. After experimental infection of 3- to 4-week-old, colostrum-deprived (seronegative) calves, the virus was detected from days 3 to 8 post-initial inoculation day (PID). An immune response was first detected 8 to 10 days PID, when BRSV-specific IgM and IgA appeared nearly simultaneously in serum, secretions, and feces. BRSV-specific IgG1 appeared only in serum on days 13 to 17 PID, and IgG2 was first detected in sera from 1 to 3 months PID. Specific IgM and IgA were detectable in the different samples for various periods. In the respiratory and eye secretions, IgA usually remained detectable for long periods, that is, for up to 3.5 months or longer. In lung lavage samples, BRSV-specific IgG1 was only incidentally demonstrated and appeared to be blood derived. The immune response of a 5-month-old calf strongly resembled that of the 3- to 4-week-old calves (feces excepted), indicating that an age effect on the immune response to BRSV is unlikely. After experimental infection of colostrum-fed, seropositive calves, both local and systemic antibody responses were largely or totally suppressed. The degree of suppression seemed to be related to the level of preinoculation virus-specific serum IgG1. Of all isotypes, IgM was least affected. Colostrum-fed animals shed virus in about equal amounts and for the same length of time as colostrum-deprived calves. Clinical signs were mild in both groups. After reinfection, no virus shedding was detected in either colostrum-deprived or colostrum-fed calves. In both groups, a secondary immune response developed, characterized by strong and rapid (from about day 6 PID) mucosal and systemic IgA responses, but reaching higher titers in colostrum-deprived calves. Also, strong mucosal, but not serum, IgM responses were observed, which, however, did not develop faster than those observed after primary infection. Naturally infected calves, showing severe signs of respiratory disease, had various levels of, most likely, maternally derived antibodies on the first day of illness. Mucosal and systemic antibody responses of various heights and durations were observed, but in general these responses were stronger than those observed after experimental infection. The results point to an important role for local IgA, rather for serum IgG1, in the protection against BRSV infection. The capacity to mount a local memory IgA response seems especially important. Priming for such a mucosal memory response is possible even when the primary immune response is severely suppressed because of the presence of material antibodies.