Live attenuated, cold-adapted (ca) monovalent and bivalent influenza A vaccines were evaluated in seronegative infants (ages 6–18 months) in a double-blind placebo-controlled trial to assess safety and immunogenicity. A total of 182 seronegative subjects received a single intranasal dose (106.2 TCID50) of ca A/Kawasaki/9/86 (H1N1) or ca A/LosAngeles/2/87 (H3N2), both as a bivalent vaccine, or placebo. Respiratory and systemic symptoms did not differ between groups after vaccination. Hemagglutination antibody seroconversions (⩾1:8)to H3N2 exceeded 90%. In contrast, seroconversions to A/Kawasaki/9/86 (HINl) were significantly less frequent in bivalent ca vaccine recipients (31%) than in monovalent ca H1N1 recipients (83%) (P < .002). During a subsequent H3N2 epidemic, nasal washes were cultured for viruses from any subject with respiratory illness. H3N2 infections documented by virus isolation were reduced by 65% in ca H3N2 recipients compared with placebo or ca H1N1 recipients (P = .01).