Clinical trials in young children of viral respiratory vaccines provide a model for determining viral and host factors in the control of primary infection. An attenuated parainfluenza 3 (para 3) vaccine was given intranasally (104-5 TCID50/dose) to 16 children ages 13 to 35 months-with 5 additional placebo inoculated controls studied for transmission and intercurrent illness. Prior natural para 3, as judged by prevaccination serum antibody, completely inhibited vaccine virus shedding and only 1/7 sero-positive children had an antibody rise. In contrast, 7/9 sero-negative children shed virus for up to 11 days after inoculation. All seronegatives had serum antibody responses by CF, HAI or plaque-neutralization tests. No nasal antibody responses were observed in 9 children (4 of whom had a serum antibody rise) tested. Mild afebrile respiratory illness was observed at the time of peak virus shedding, 8-9 days post inoculation, suggesting illness was a function of virus replication. Thus, multiple laboratory passages (4 in monkey kidney, 75 in eggs) did not sufficiently attenuate para 3 for seronegative children. In spite of symptomatic shedding in vaccinees no transmission to controls was observed. The duration of para 3 vaccine shedding in seronegatives was uniform with a mean of 9.4 days. This is similiar to shedding by seronegatives of attenuated respiratory syncytial virus, 10 days, and influenza, 8.6 days; and suggests a common host mechanism for terminating primary respiratory viral infection.